{"id": "scene0000_00_0", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the coffee kettle appear?", "choices": null, "answer": "No", "target": "coffee kettle", "split": "train"}
{"id": "scene0000_00_1", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the candle appear?", "choices": null, "answer": "No", "target": "candle", "split": "train"}
{"id": "scene0000_00_2", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0000_00_3", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the guitar in this scene?\nA. backpack\nB. cabinet\nC. scale\nD. clock", "choices": ["A. backpack", "B. cabinet", "C. scale", "D. clock"], "answer": "C. scale", "target": "guitar", "split": "train"}
{"id": "scene0000_00_4", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the guitar in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The guitar is to the side of and above the scale, about 38 cm away.", "target": "guitar", "split": "train"}
{"id": "scene0000_00_5", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the guitar in structured JSON format.", "choices": null, "answer": "{\"target\": \"guitar\", \"support_surface\": \"scale\", \"references\": [{\"object\": \"scale\", \"relation\": \"side_above\", \"distance_cm\": 37.8}]}", "target": "guitar", "split": "train"}
{"id": "scene0000_00_6", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "Yes", "target": "scale", "split": "train"}
{"id": "scene0000_00_7", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scale in this scene?\nA. guitar\nB. laundry basket\nC. backpack\nD. cabinet", "choices": ["A. guitar", "B. laundry basket", "C. backpack", "D. cabinet"], "answer": "A. guitar", "target": "scale", "split": "train"}
{"id": "scene0000_00_8", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is to the side of and below the guitar, about 38 cm away. It is also to the side of and below the laundry basket, about 53 cm away.", "target": "scale", "split": "train"}
{"id": "scene0000_00_9", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"laundry basket\", \"references\": [{\"object\": \"guitar\", \"relation\": \"side_below\", \"distance_cm\": 37.8}, {\"object\": \"laundry basket\", \"relation\": \"side_below\", \"distance_cm\": 53.0}]}", "target": "scale", "split": "train"}
{"id": "scene0000_00_10", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "scene0000_00_11", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. pillow\nB. backpack\nC. cabinet\nD. trash can", "choices": ["A. pillow", "B. backpack", "C. cabinet", "D. trash can"], "answer": "A. pillow", "target": "tissue box", "split": "train"}
{"id": "scene0000_00_12", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the nightstand. It is also to the side of and below the pillow, about 37 cm away. It is further to the side of and above the trash can, about 46 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0000_00_13", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 37.2}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 45.7}]}", "target": "tissue box", "split": "train"}
{"id": "scene0000_00_14", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "Yes", "target": "toaster", "split": "train"}
{"id": "scene0000_00_15", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toaster in this scene?\nA. microwave\nB. kitchen cabinets\nC. cabinet\nD. dish rack", "choices": ["A. microwave", "B. kitchen cabinets", "C. cabinet", "D. dish rack"], "answer": "D. dish rack", "target": "toaster", "split": "train"}
{"id": "scene0000_00_16", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toaster in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster is near the dish rack, about 16 cm away. It is also to the side of and below the kitchen cabinets, about 27 cm away. It is further to the side of and below the microwave, about 37 cm away.", "target": "toaster", "split": "train"}
{"id": "scene0000_00_17", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toaster in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"dish rack\", \"relation\": \"near\", \"distance_cm\": 15.7}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_below\", \"distance_cm\": 27.1}]}", "target": "toaster", "split": "train"}
{"id": "scene0000_00_18", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "18", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "train"}
{"id": "scene0000_00_19", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "19", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toaster oven in this scene?\nA. clock\nB. microwave\nC. backpack\nD. cabinet", "choices": ["A. clock", "B. microwave", "C. backpack", "D. cabinet"], "answer": "D. cabinet", "target": "toaster oven", "split": "train"}
{"id": "scene0000_00_20", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "20", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toaster oven in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster oven is to the side of and above the cabinet, about 13 cm away. It is also to the side of and below the microwave, about 28 cm away.", "target": "toaster oven", "split": "train"}
{"id": "scene0000_00_21", "scene_id": "scene0000_00", "source_dataset": "scannet200", "local_sample_id": "21", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toaster oven in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster oven\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 13.0}, {\"object\": \"microwave\", \"relation\": \"side_below\", \"distance_cm\": 28.2}]}", "target": "toaster oven", "split": "train"}
{"id": "scene0002_00_0", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0002_00_1", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. fan\nB. fireplace\nC. radiator\nD. table", "choices": ["A. fan", "B. fireplace", "C. radiator", "D. table"], "answer": "B. fireplace", "target": "bag", "split": "train"}
{"id": "scene0002_00_2", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the fireplace, about 8 cm away. It is also to the side of and below the table, about 15 cm away. It is further to the side of and below the radiator, about 31 cm away.", "target": "bag", "split": "train"}
{"id": "scene0002_00_3", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"table\", \"references\": [{\"object\": \"fireplace\", \"relation\": \"side_below\", \"distance_cm\": 8.3}, {\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 15.4}]}", "target": "bag", "split": "train"}
{"id": "scene0002_00_4", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0002_00_5", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. closet\nB. hat\nC. chair\nD. pillow", "choices": ["A. closet", "B. hat", "C. chair", "D. pillow"], "answer": "B. hat", "target": "bottle", "split": "train"}
{"id": "scene0002_00_6", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the cabinet. It is also next to the hat, about 4 cm away. It is further to the side of and below the closet, about 49 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0002_00_7", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"hat\", \"relation\": \"next_to\", \"distance_cm\": 3.5}, {\"object\": \"closet\", \"relation\": \"side_below\", \"distance_cm\": 49.4}]}", "target": "bottle", "split": "train"}
{"id": "scene0002_00_8", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "Yes", "target": "hat", "split": "train"}
{"id": "scene0002_00_9", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hat in this scene?\nA. pillow\nB. closet\nC. chair\nD. bottle", "choices": ["A. pillow", "B. closet", "C. chair", "D. bottle"], "answer": "D. bottle", "target": "hat", "split": "train"}
{"id": "scene0002_00_10", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hat in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hat is on the cabinet. It is also under the bottle, about 4 cm away. It is further to the side of and below the closet, about 39 cm away.", "target": "hat", "split": "train"}
{"id": "scene0002_00_11", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hat in structured JSON format.", "choices": null, "answer": "{\"target\": \"hat\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"bottle\", \"relation\": \"under\", \"distance_cm\": 3.5}, {\"object\": \"closet\", \"relation\": \"side_below\", \"distance_cm\": 39.0}]}", "target": "hat", "split": "train"}
{"id": "scene0002_00_12", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the shoe appear?", "choices": null, "answer": "Yes", "target": "shoe", "split": "train"}
{"id": "scene0002_00_13", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the shoe in this scene?\nA. fan\nB. controller\nC. radiator\nD. chair", "choices": ["A. fan", "B. controller", "C. radiator", "D. chair"], "answer": "C. radiator", "target": "shoe", "split": "train"}
{"id": "scene0002_00_14", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the shoe in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The shoe is to the side of and below the radiator, about 11 cm away. It is also to the side of and below the chair, about 39 cm away. It is further to the side of and below the controller, about 49 cm away.", "target": "shoe", "split": "train"}
{"id": "scene0002_00_15", "scene_id": "scene0002_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the shoe in structured JSON format.", "choices": null, "answer": "{\"target\": \"shoe\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"radiator\", \"relation\": \"side_below\", \"distance_cm\": 10.5}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 39.1}]}", "target": "shoe", "split": "train"}
{"id": "scene0003_00_0", "scene_id": "scene0003_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0003_00_1", "scene_id": "scene0003_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. cabinet\nB. sink\nC. paper towel dispenser\nD. kitchen counter", "choices": ["A. cabinet", "B. sink", "C. paper towel dispenser", "D. kitchen counter"], "answer": "D. kitchen counter", "target": "soap dispenser", "split": "train"}
{"id": "scene0003_00_2", "scene_id": "scene0003_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the kitchen counter, about 12 cm away. It is also to the side of and above the sink, about 28 cm away. It is further to the side of and below the paper towel dispenser, about 44 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0003_00_3", "scene_id": "scene0003_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 11.6}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 27.9}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0010_00_0", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "scene0010_00_1", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. chair\nB. desk\nC. power strip\nD. speaker", "choices": ["A. chair", "B. desk", "C. power strip", "D. speaker"], "answer": "B. desk", "target": "bag", "split": "train"}
{"id": "scene0010_00_2", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the desk, about 6 cm away. It is also to the side of and below the chair, about 24 cm away. It is further to the side of and below the speaker, about 51 cm away.", "target": "bag", "split": "train"}
{"id": "scene0010_00_3", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 24.0}]}", "target": "bag", "split": "train"}
{"id": "scene0010_00_4", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "train"}
{"id": "scene0010_00_5", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. speaker\nB. file cabinet\nC. monitor\nD. computer tower", "choices": ["A. speaker", "B. file cabinet", "C. monitor", "D. computer tower"], "answer": "A. speaker", "target": "book", "split": "train"}
{"id": "scene0010_00_6", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is near the speaker, about 15 cm away. It is also to the side of and above the file cabinet, about 18 cm away. It is further to the side of and below the monitor, about 31 cm away.", "target": "book", "split": "train"}
{"id": "scene0010_00_7", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"speaker\", \"relation\": \"near\", \"distance_cm\": 15.2}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 17.7}]}", "target": "book", "split": "train"}
{"id": "scene0010_00_8", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0010_00_9", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. box\nB. chair\nC. computer tower\nD. monitor", "choices": ["A. box", "B. chair", "C. computer tower", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0010_00_10", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the monitor, about 26 cm away. It is further to the side of and below the box, about 33 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0010_00_11", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 25.6}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 32.7}]}", "target": "keyboard", "split": "train"}
{"id": "scene0010_00_12", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0010_00_13", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. file cabinet\nB. books\nC. chair\nD. computer tower", "choices": ["A. file cabinet", "B. books", "C. chair", "D. computer tower"], "answer": "B. books", "target": "laptop", "split": "train"}
{"id": "scene0010_00_14", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is to the side of and below the books, about 9 cm away. It is also to the side of and above the file cabinet, about 12 cm away. It is further to the side of and above the chair, about 13 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0010_00_15", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 9.0}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 11.5}]}", "target": "laptop", "split": "train"}
{"id": "scene0010_00_16", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "scene0010_00_17", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor\nB. speaker\nC. computer tower\nD. bag", "choices": ["A. monitor", "B. speaker", "C. computer tower", "D. bag"], "answer": "C. computer tower", "target": "power strip", "split": "train"}
{"id": "scene0010_00_18", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the computer tower, about 26 cm away. It is also to the side of and below the bag, about 58 cm away. It is further to the side of and below the speaker, about 70 cm away.", "target": "power strip", "split": "train"}
{"id": "scene0010_00_19", "scene_id": "scene0010_00", "source_dataset": "scannet200", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_below\", \"distance_cm\": 25.5}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 57.8}]}", "target": "power strip", "split": "train"}
{"id": "scene0012_00_0", "scene_id": "scene0012_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plunger appear?", "choices": null, "answer": "No", "target": "plunger", "split": "train"}
{"id": "scene0014_00_0", "scene_id": "scene0014_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plunger appear?", "choices": null, "answer": "No", "target": "plunger", "split": "train"}
{"id": "scene0014_00_1", "scene_id": "scene0014_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. bathroom cabinet\nB. sink\nC. toilet paper\nD. monitor", "choices": ["A. bathroom cabinet", "B. sink", "C. toilet paper", "D. monitor"], "answer": "B. sink", "target": "soap dispenser", "split": "train"}
{"id": "scene0014_00_2", "scene_id": "scene0014_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the sink, about 38 cm away. It is also to the side of and above the bathroom cabinet, about 42 cm away. It is further to the side of and above the toilet paper, about 65 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0014_00_3", "scene_id": "scene0014_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 38.3}, {\"object\": \"bathroom cabinet\", \"relation\": \"side_above\", \"distance_cm\": 42.5}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0015_00_0", "scene_id": "scene0015_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0015_00_1", "scene_id": "scene0015_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. table\nB. chair\nC. monitor\nD. water bottle", "choices": ["A. table", "B. chair", "C. monitor", "D. water bottle"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0015_00_2", "scene_id": "scene0015_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "train"}
{"id": "scene0015_00_3", "scene_id": "scene0015_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. backpack\nB. monitor\nC. chair\nD. table", "choices": ["A. backpack", "B. monitor", "C. chair", "D. table"], "answer": "C. chair", "target": "water bottle", "split": "train"}
{"id": "scene0015_00_4", "scene_id": "scene0015_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also to the side of and above the chair, about 53 cm away.", "target": "water bottle", "split": "train"}
{"id": "scene0015_00_5", "scene_id": "scene0015_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 53.2}]}", "target": "water bottle", "split": "train"}
{"id": "scene0016_00_0", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "train"}
{"id": "scene0016_00_1", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. pillow\nB. storage container\nC. microwave\nD. shelf", "choices": ["A. pillow", "B. storage container", "C. microwave", "D. shelf"], "answer": "D. shelf", "target": "bag", "split": "train"}
{"id": "scene0016_00_2", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the shelf, about 6 cm away. It is also to the side of and below the microwave, about 17 cm away. It is further to the side of and above the pillow, about 22 cm away.", "target": "bag", "split": "train"}
{"id": "scene0016_00_3", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"microwave\", \"relation\": \"side_below\", \"distance_cm\": 17.4}]}", "target": "bag", "split": "train"}
{"id": "scene0016_00_4", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "Yes", "target": "ball", "split": "train"}
{"id": "scene0016_00_5", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. piano bench\nB. piano\nC. bag\nD. shoes", "choices": ["A. piano bench", "B. piano", "C. bag", "D. shoes"], "answer": "B. piano", "target": "ball", "split": "train"}
{"id": "scene0016_00_6", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is to the side of and below the piano, about 5 cm away. It is also near the shoes, about 19 cm away. It is further to the side of and below the piano bench, about 44 cm away.", "target": "ball", "split": "train"}
{"id": "scene0016_00_7", "scene_id": "scene0016_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"piano bench\", \"references\": [{\"object\": \"piano\", \"relation\": \"side_below\", \"distance_cm\": 5.1}, {\"object\": \"shoes\", \"relation\": \"near\", \"distance_cm\": 19.0}]}", "target": "ball", "split": "train"}
{"id": "scene0017_00_0", "scene_id": "scene0017_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0017_00_1", "scene_id": "scene0017_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. file cabinet\nB. monitor\nC. keyboard\nD. windowsill", "choices": ["A. file cabinet", "B. monitor", "C. keyboard", "D. windowsill"], "answer": "B. monitor", "target": "book", "split": "train"}
{"id": "scene0017_00_2", "scene_id": "scene0017_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is to the side of and below the monitor, about 23 cm away. It is also near the keyboard, about 30 cm away. It is further to the side of and above the file cabinet, about 36 cm away.", "target": "book", "split": "train"}
{"id": "scene0017_00_3", "scene_id": "scene0017_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 22.8}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 29.7}]}", "target": "book", "split": "train"}
{"id": "scene0018_00_0", "scene_id": "scene0018_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "Yes", "target": "paper cutter", "split": "train"}
{"id": "scene0018_00_1", "scene_id": "scene0018_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper cutter in this scene?\nA. cabinets\nB. battery disposal jar\nC. power strip\nD. monitor", "choices": ["A. cabinets", "B. battery disposal jar", "C. power strip", "D. monitor"], "answer": "B. battery disposal jar", "target": "paper cutter", "split": "train"}
{"id": "scene0018_00_2", "scene_id": "scene0018_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper cutter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper cutter is to the side of and below the battery disposal jar, about 7 cm away.", "target": "paper cutter", "split": "train"}
{"id": "scene0018_00_3", "scene_id": "scene0018_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper cutter in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper cutter\", \"support_surface\": \"cabinets\", \"references\": [{\"object\": \"battery disposal jar\", \"relation\": \"side_below\", \"distance_cm\": 7.5}]}", "target": "paper cutter", "split": "train"}
{"id": "scene0018_00_4", "scene_id": "scene0018_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the candle appear?", "choices": null, "answer": "No", "target": "candle", "split": "train"}
{"id": "scene0020_00_0", "scene_id": "scene0020_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0020_00_1", "scene_id": "scene0020_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. book\nB. monitor\nC. chair\nD. desk", "choices": ["A. book", "B. monitor", "C. chair", "D. desk"], "answer": "D. desk", "target": "backpack", "split": "train"}
{"id": "scene0020_00_2", "scene_id": "scene0020_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is near the desk, about 19 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0020_00_3", "scene_id": "scene0020_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"near\", \"distance_cm\": 18.7}]}", "target": "backpack", "split": "train"}
{"id": "scene0030_00_0", "scene_id": "scene0030_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cd case appear?", "choices": null, "answer": "No", "target": "cd case", "split": "train"}
{"id": "scene0030_00_1", "scene_id": "scene0030_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. box\nB. table\nC. blackboard\nD. clock", "choices": ["A. box", "B. table", "C. blackboard", "D. clock"], "answer": "A. box", "target": "bag", "split": "train"}
{"id": "scene0030_00_2", "scene_id": "scene0030_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "Yes", "target": "clock", "split": "train"}
{"id": "scene0041_00_0", "scene_id": "scene0041_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dustpan appear?", "choices": null, "answer": "No", "target": "dustpan", "split": "train"}
{"id": "scene0041_00_1", "scene_id": "scene0041_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. table\nB. chair\nC. desk\nD. monitor", "choices": ["A. table", "B. chair", "C. desk", "D. monitor"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0041_00_2", "scene_id": "scene0041_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0043_00_0", "scene_id": "scene0043_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0043_00_1", "scene_id": "scene0043_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bench\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0044_00_0", "scene_id": "scene0044_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0046_00_0", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the coffee kettle appear?", "choices": null, "answer": "No", "target": "coffee kettle", "split": "train"}
{"id": "scene0046_00_1", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. clothes\nB. luggage stand\nC. backpack\nD. chair", "choices": ["A. clothes", "B. luggage stand", "C. backpack", "D. chair"], "answer": "B. luggage stand", "target": "laptop", "split": "train"}
{"id": "scene0046_00_2", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also to the side of and above the luggage stand, about 3 cm away. It is further to the side of and above the chair, about 21 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0046_00_3", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"luggage stand\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 21.1}]}", "target": "laptop", "split": "train"}
{"id": "scene0046_00_4", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0046_00_5", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. toilet flush button\nB. toilet\nC. pillow\nD. light", "choices": ["A. toilet flush button", "B. toilet", "C. pillow", "D. light"], "answer": "A. toilet flush button", "target": "toilet paper", "split": "train"}
{"id": "scene0046_00_6", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet flush button, about 15 cm away. It is also to the side of and above the toilet, about 30 cm away. It is further to the side of and below the light, about 47 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0046_00_7", "scene_id": "scene0046_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"toilet flush button\", \"relation\": \"side_above\", \"distance_cm\": 15.2}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 30.1}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0049_00_0", "scene_id": "scene0049_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0049_00_1", "scene_id": "scene0049_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the clock in this scene?\nA. box\nB. monitor\nC. shelf\nD. chair", "choices": ["A. box", "B. monitor", "C. shelf", "D. chair"], "answer": "C. shelf", "target": "clock", "split": "train"}
{"id": "scene0049_00_2", "scene_id": "scene0049_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The clock is to the side of and above the shelf, about 11 cm away. It is also to the side of and above the box, about 14 cm away.", "target": "clock", "split": "train"}
{"id": "scene0049_00_3", "scene_id": "scene0049_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"clock\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 10.6}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 13.6}]}", "target": "clock", "split": "train"}
{"id": "scene0050_00_0", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0050_00_1", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. shelf\nB. cup\nC. laptop\nD. desk", "choices": ["A. shelf", "B. cup", "C. laptop", "D. desk"], "answer": "D. desk", "target": "bag", "split": "train"}
{"id": "scene0050_00_2", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the desk, about 14 cm away. It is also to the side of and below the shelf, about 16 cm away. It is further to the side of and above the cup, about 34 cm away.", "target": "bag", "split": "train"}
{"id": "scene0050_00_3", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 13.7}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 16.0}]}", "target": "bag", "split": "train"}
{"id": "scene0050_00_4", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "Yes", "target": "clock", "split": "train"}
{"id": "scene0050_00_5", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0050_00_6", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. backpack\nB. laptop\nC. bag\nD. shelf", "choices": ["A. backpack", "B. laptop", "C. bag", "D. shelf"], "answer": "B. laptop", "target": "cup", "split": "train"}
{"id": "scene0050_00_7", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also near the laptop, about 25 cm away. It is further to the side of and below the bag, about 34 cm away.", "target": "cup", "split": "train"}
{"id": "scene0050_00_8", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 25.1}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 33.5}]}", "target": "cup", "split": "train"}
{"id": "scene0050_00_9", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "scene0050_00_10", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the ottoman.", "target": "tissue box", "split": "train"}
{"id": "scene0050_00_11", "scene_id": "scene0050_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"ottoman\", \"references\": []}", "target": "tissue box", "split": "train"}
{"id": "scene0051_00_0", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0051_00_1", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. chair\nB. mouse\nC. dresser\nD. monitor", "choices": ["A. chair", "B. mouse", "C. dresser", "D. monitor"], "answer": "D. monitor", "target": "laptop", "split": "train"}
{"id": "scene0051_00_2", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the table. It is also to the side of and below the monitor, about 9 cm away. It is further to the side of and above the chair, about 41 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0051_00_3", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 40.9}]}", "target": "laptop", "split": "train"}
{"id": "scene0051_00_4", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "scene0051_00_5", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. monitor\nC. fan\nD. computer tower", "choices": ["A. chair", "B. monitor", "C. fan", "D. computer tower"], "answer": "B. monitor", "target": "mouse", "split": "train"}
{"id": "scene0051_00_6", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and below the monitor, about 17 cm away. It is further to the side of and above the chair, about 43 cm away.", "target": "mouse", "split": "train"}
{"id": "scene0051_00_7", "scene_id": "scene0051_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 17.0}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 43.0}]}", "target": "mouse", "split": "train"}
{"id": "scene0054_00_0", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0054_00_1", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. carseat\nB. dolly\nC. crate\nD. shelf", "choices": ["A. carseat", "B. dolly", "C. crate", "D. shelf"], "answer": "A. carseat", "target": "bag", "split": "train"}
{"id": "scene0054_00_2", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the carseat, about 15 cm away. It is also to the side of and below the crate, about 21 cm away. It is further to the side of and below the shelf, about 39 cm away.", "target": "bag", "split": "train"}
{"id": "scene0054_00_3", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"carseat\", \"relation\": \"side_below\", \"distance_cm\": 14.7}, {\"object\": \"crate\", \"relation\": \"side_below\", \"distance_cm\": 21.4}]}", "target": "bag", "split": "train"}
{"id": "scene0054_00_4", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "Yes", "target": "ball", "split": "train"}
{"id": "scene0054_00_5", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. book\nB. bathroom vanity\nC. bag\nD. stool", "choices": ["A. book", "B. bathroom vanity", "C. bag", "D. stool"], "answer": "D. stool", "target": "ball", "split": "train"}
{"id": "scene0054_00_6", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is next to the stool, about 14 cm away.", "target": "ball", "split": "train"}
{"id": "scene0054_00_7", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"stool\", \"relation\": \"next_to\", \"distance_cm\": 14.1}]}", "target": "ball", "split": "train"}
{"id": "scene0054_00_8", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0054_00_9", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. bag\nB. bathroom vanity\nC. table\nD. ball", "choices": ["A. bag", "B. bathroom vanity", "C. table", "D. ball"], "answer": "C. table", "target": "book", "split": "train"}
{"id": "scene0054_00_10", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the table. It is also near the bag, about 30 cm away.", "target": "book", "split": "train"}
{"id": "scene0054_00_11", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bag\", \"relation\": \"near\", \"distance_cm\": 29.5}]}", "target": "book", "split": "train"}
{"id": "scene0054_00_12", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0054_00_13", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. toilet\nB. sink\nC. tissue box\nD. bathroom vanity", "choices": ["A. toilet", "B. sink", "C. tissue box", "D. bathroom vanity"], "answer": "D. bathroom vanity", "target": "soap dish", "split": "train"}
{"id": "scene0054_00_14", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the bathroom vanity, about 8 cm away. It is also to the side of and above the sink, about 9 cm away. It is further to the side of and above the toilet, about 37 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0054_00_15", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"tissue box\", \"references\": [{\"object\": \"bathroom vanity\", \"relation\": \"side_above\", \"distance_cm\": 7.5}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 8.7}]}", "target": "soap dish", "split": "train"}
{"id": "scene0054_00_16", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "scene0054_00_17", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. box\nB. soap dish\nC. sink\nD. bookshelf", "choices": ["A. box", "B. soap dish", "C. sink", "D. bookshelf"], "answer": "C. sink", "target": "tissue box", "split": "train"}
{"id": "scene0054_00_18", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and above the sink, about 11 cm away. It is also to the side of and below the bookshelf, about 30 cm away. It is further to the side of and below the soap dish, about 43 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0054_00_19", "scene_id": "scene0054_00", "source_dataset": "scannet200", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 11.4}, {\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 29.7}]}", "target": "tissue box", "split": "train"}
{"id": "scene0058_00_0", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "train"}
{"id": "scene0058_00_1", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. chair\nC. computer tower\nD. desk", "choices": ["A. monitor", "B. chair", "C. computer tower", "D. desk"], "answer": "A. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0058_00_2", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the monitor, about 10 cm away. It is also to the side of and above the chair, about 34 cm away. It is further to the side of and above the computer tower, about 34 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0058_00_3", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 33.8}]}", "target": "keyboard", "split": "train"}
{"id": "scene0058_00_4", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "scene0058_00_5", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. desk\nB. fan\nC. computer tower\nD. chair", "choices": ["A. desk", "B. fan", "C. computer tower", "D. chair"], "answer": "B. fan", "target": "power strip", "split": "train"}
{"id": "scene0058_00_6", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the fan, about 31 cm away. It is also to the side of and above the chair, about 53 cm away.", "target": "power strip", "split": "train"}
{"id": "scene0058_00_7", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"fan\", \"relation\": \"side_below\", \"distance_cm\": 30.8}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 52.8}]}", "target": "power strip", "split": "train"}
{"id": "scene0058_00_8", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0058_00_9", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "train"}
{"id": "scene0058_00_10", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. table\nC. plant\nD. keyboard", "choices": ["A. monitor", "B. table", "C. plant", "D. keyboard"], "answer": "B. table", "target": "telephone", "split": "train"}
{"id": "scene0058_00_11", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also to the side of and above the table, about 8 cm away. It is further to the side of and below the monitor, about 14 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0058_00_12", "scene_id": "scene0058_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 8.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 14.4}]}", "target": "telephone", "split": "train"}
{"id": "scene0059_00_0", "scene_id": "scene0059_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0059_00_1", "scene_id": "scene0059_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. doors\nB. armchair\nC. chair\nD. bag", "choices": ["A. doors", "B. armchair", "C. chair", "D. bag"], "answer": "B. armchair", "target": "backpack", "split": "train"}
{"id": "scene0059_00_2", "scene_id": "scene0059_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the armchair, about 9 cm away. It is also to the side of and below the doors, about 17 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0059_00_3", "scene_id": "scene0059_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"armchair\", \"relation\": \"side_above\", \"distance_cm\": 9.1}, {\"object\": \"doors\", \"relation\": \"side_below\", \"distance_cm\": 16.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0059_00_4", "scene_id": "scene0059_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0059_00_5", "scene_id": "scene0059_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. chair\nB. doors\nC. backpack\nD. armchair", "choices": ["A. chair", "B. doors", "C. backpack", "D. armchair"], "answer": "A. chair", "target": "bag", "split": "train"}
{"id": "scene0060_00_0", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0060_00_1", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cabinets\nB. machine\nC. monitor\nD. paper cutter", "choices": ["A. cabinets", "B. machine", "C. monitor", "D. paper cutter"], "answer": "B. machine", "target": "backpack", "split": "train"}
{"id": "scene0060_00_2", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the machine, about 36 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0060_00_3", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"machine\", \"references\": [{\"object\": \"machine\", \"relation\": \"side_below\", \"distance_cm\": 35.9}]}", "target": "backpack", "split": "train"}
{"id": "scene0060_00_4", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "scene0060_00_5", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper cutter in this scene?\nA. machine\nB. monitor\nC. backpack\nD. cabinets", "choices": ["A. machine", "B. monitor", "C. backpack", "D. cabinets"], "answer": "B. monitor", "target": "paper cutter", "split": "train"}
{"id": "scene0060_00_6", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper cutter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper cutter is to the side of and below the monitor, about 25 cm away.", "target": "paper cutter", "split": "train"}
{"id": "scene0060_00_7", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper cutter in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper cutter\", \"support_surface\": \"cabinets\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 24.8}]}", "target": "paper cutter", "split": "train"}
{"id": "scene0060_00_8", "scene_id": "scene0060_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0062_00_0", "scene_id": "scene0062_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "scene0062_00_1", "scene_id": "scene0062_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. paper towel dispenser\nB. trash can\nC. sink\nD. mirror", "choices": ["A. paper towel dispenser", "B. trash can", "C. sink", "D. mirror"], "answer": "D. mirror", "target": "soap dispenser", "split": "train"}
{"id": "scene0062_00_2", "scene_id": "scene0062_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is on the counter. It is also next to the mirror, about 4 cm away. It is further next to the paper towel dispenser, about 9 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0062_00_3", "scene_id": "scene0062_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"mirror\", \"relation\": \"next_to\", \"distance_cm\": 4.2}, {\"object\": \"paper towel dispenser\", \"relation\": \"next_to\", \"distance_cm\": 9.4}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0064_00_0", "scene_id": "scene0064_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "train"}
{"id": "scene0064_00_1", "scene_id": "scene0064_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. coffee table\nB. chair\nC. couch\nD. monitor", "choices": ["A. coffee table", "B. chair", "C. couch", "D. monitor"], "answer": "A. coffee table", "target": "backpack", "split": "train"}
{"id": "scene0064_00_2", "scene_id": "scene0064_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the coffee table, about 25 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0064_00_3", "scene_id": "scene0064_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"couch\", \"references\": [{\"object\": \"coffee table\", \"relation\": \"side_above\", \"distance_cm\": 25.0}]}", "target": "backpack", "split": "train"}
{"id": "scene0065_00_0", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0065_00_1", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plunger in this scene?\nA. toilet\nB. mirror\nC. sink\nD. bathroom cabinet", "choices": ["A. toilet", "B. mirror", "C. sink", "D. bathroom cabinet"], "answer": "A. toilet", "target": "plunger", "split": "train"}
{"id": "scene0065_00_2", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plunger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plunger is to the side of and below the toilet, about 57 cm away.", "target": "plunger", "split": "train"}
{"id": "scene0065_00_3", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plunger in structured JSON format.", "choices": null, "answer": "{\"target\": \"plunger\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 56.8}]}", "target": "plunger", "split": "train"}
{"id": "scene0065_00_4", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0065_00_5", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bathroom cabinet\nB. sink\nC. mirror\nD. toilet", "choices": ["A. bathroom cabinet", "B. sink", "C. mirror", "D. toilet"], "answer": "D. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0065_00_6", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is near the toilet, about 24 cm away. It is also to the side of and below the sink, about 33 cm away. It is further to the side of and below the mirror, about 40 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0065_00_7", "scene_id": "scene0065_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"near\", \"distance_cm\": 23.6}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 32.7}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0068_00_0", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0068_00_1", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the fire extinguisher in this scene?\nA. trash bag\nB. coffee box\nC. bottle\nD. cup", "choices": ["A. trash bag", "B. coffee box", "C. bottle", "D. cup"], "answer": "A. trash bag", "target": "fire extinguisher", "split": "train"}
{"id": "scene0068_00_2", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the fire extinguisher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The fire extinguisher is to the side of and above the trash bag, about 44 cm away.", "target": "fire extinguisher", "split": "train"}
{"id": "scene0068_00_3", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the fire extinguisher in structured JSON format.", "choices": null, "answer": "{\"target\": \"fire extinguisher\", \"support_surface\": \"trash bag\", \"references\": [{\"object\": \"trash bag\", \"relation\": \"side_above\", \"distance_cm\": 43.7}]}", "target": "fire extinguisher", "split": "train"}
{"id": "scene0068_00_4", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "Yes", "target": "toaster oven", "split": "train"}
{"id": "scene0068_00_5", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toaster oven in this scene?\nA. coffee box\nB. cup\nC. bottle\nD. dishwasher", "choices": ["A. coffee box", "B. cup", "C. bottle", "D. dishwasher"], "answer": "C. bottle", "target": "toaster oven", "split": "train"}
{"id": "scene0068_00_6", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toaster oven in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster oven is next to the bottle, about 11 cm away. It is also near the cup, about 21 cm away. It is further near the coffee box, about 27 cm away.", "target": "toaster oven", "split": "train"}
{"id": "scene0068_00_7", "scene_id": "scene0068_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toaster oven in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster oven\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 10.5}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 20.8}]}", "target": "toaster oven", "split": "train"}
{"id": "scene0073_00_0", "scene_id": "scene0073_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "train"}
{"id": "scene0073_00_1", "scene_id": "scene0073_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the dumbbell in structured JSON format.", "choices": null, "answer": "{\"target\": \"dumbbell\", \"support_surface\": \"bed\", \"references\": []}", "target": "dumbbell", "split": "train"}
{"id": "scene0074_00_0", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0074_00_1", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. pillow\nB. shelf\nC. costume\nD. clothing", "choices": ["A. pillow", "B. shelf", "C. costume", "D. clothing"], "answer": "B. shelf", "target": "bag", "split": "train"}
{"id": "scene0074_00_2", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the shelf, about 8 cm away. It is also above the clothing, about 14 cm away. It is further to the side of and above the pillow, about 18 cm away.", "target": "bag", "split": "train"}
{"id": "scene0074_00_3", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 7.6}, {\"object\": \"clothing\", \"relation\": \"above\", \"distance_cm\": 13.7}]}", "target": "bag", "split": "train"}
{"id": "scene0074_00_4", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "Yes", "target": "hat", "split": "train"}
{"id": "scene0074_00_5", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hat in this scene?\nA. stuffed animal\nB. box\nC. stool\nD. bag", "choices": ["A. stuffed animal", "B. box", "C. stool", "D. bag"], "answer": "A. stuffed animal", "target": "hat", "split": "train"}
{"id": "scene0074_00_6", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hat in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hat is to the side of and below the stuffed animal, about 37 cm away. It is also to the side of and below the stool, about 57 cm away.", "target": "hat", "split": "train"}
{"id": "scene0074_00_7", "scene_id": "scene0074_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hat in structured JSON format.", "choices": null, "answer": "{\"target\": \"hat\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"stuffed animal\", \"relation\": \"side_below\", \"distance_cm\": 36.9}, {\"object\": \"stool\", \"relation\": \"side_below\", \"distance_cm\": 56.9}]}", "target": "hat", "split": "train"}
{"id": "scene0076_00_0", "scene_id": "scene0076_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "No", "target": "hair dryer", "split": "train"}
{"id": "scene0076_00_1", "scene_id": "scene0076_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. kitchen counter\nB. paper towel dispenser\nC. sink\nD. kitchen cabinets", "choices": ["A. kitchen counter", "B. paper towel dispenser", "C. sink", "D. kitchen cabinets"], "answer": "A. kitchen counter", "target": "soap dispenser", "split": "train"}
{"id": "scene0076_00_2", "scene_id": "scene0076_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the kitchen counter. It is also to the side of and above the sink, about 13 cm away. It is further to the side of and below the paper towel dispenser, about 19 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0076_00_3", "scene_id": "scene0076_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 2.2}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 12.9}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0080_00_0", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the shoe appear?", "choices": null, "answer": "No", "target": "shoe", "split": "train"}
{"id": "scene0080_00_1", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. mirror\nB. toilet paper\nC. trash can\nD. bathroom cabinet", "choices": ["A. mirror", "B. toilet paper", "C. trash can", "D. bathroom cabinet"], "answer": "B. toilet paper", "target": "book", "split": "train"}
{"id": "scene0080_00_2", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is near the toilet paper, about 19 cm away. It is also to the side of and above the bathroom cabinet, about 26 cm away. It is further to the side of and above the trash can, about 39 cm away.", "target": "book", "split": "train"}
{"id": "scene0080_00_3", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"near\", \"distance_cm\": 19.1}, {\"object\": \"bathroom cabinet\", \"relation\": \"side_above\", \"distance_cm\": 25.7}]}", "target": "book", "split": "train"}
{"id": "scene0080_00_4", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "No", "target": "telephone", "split": "train"}
{"id": "scene0080_00_5", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. mirror\nB. sink\nC. power outlet\nD. decoration", "choices": ["A. mirror", "B. sink", "C. power outlet", "D. decoration"], "answer": "A. mirror", "target": "soap dispenser", "split": "train"}
{"id": "scene0080_00_6", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and below the mirror, about 8 cm away. It is also to the side of and above the sink, about 12 cm away. It is further to the side of and below the decoration, about 35 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0080_00_7", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 7.9}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 12.5}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0080_00_8", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0080_00_9", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. sink\nB. toilet\nC. trash can\nD. book", "choices": ["A. sink", "B. toilet", "C. trash can", "D. book"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0080_00_10", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 12 cm away. It is also near the book, about 19 cm away. It is further to the side of and below the sink, about 20 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0080_00_11", "scene_id": "scene0080_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 11.6}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 19.1}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0085_00_0", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dumbbell appear?", "choices": null, "answer": "No", "target": "dumbbell", "split": "train"}
{"id": "scene0085_00_1", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the alarm clock in this scene?\nA. lamp\nB. telephone\nC. backpack\nD. pillows", "choices": ["A. lamp", "B. telephone", "C. backpack", "D. pillows"], "answer": "B. telephone", "target": "alarm clock", "split": "train"}
{"id": "scene0085_00_2", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the alarm clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The alarm clock is on the nightstand. It is also near the telephone, about 22 cm away. It is further to the side of and below the pillows, about 52 cm away.", "target": "alarm clock", "split": "train"}
{"id": "scene0085_00_3", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the alarm clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"alarm clock\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"telephone\", \"relation\": \"near\", \"distance_cm\": 21.5}, {\"object\": \"pillows\", \"relation\": \"side_below\", \"distance_cm\": 51.7}]}", "target": "alarm clock", "split": "train"}
{"id": "scene0085_00_4", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0085_00_5", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. alarm clock\nC. radiator\nD. clothes", "choices": ["A. chair", "B. alarm clock", "C. radiator", "D. clothes"], "answer": "A. chair", "target": "backpack", "split": "train"}
{"id": "scene0085_00_6", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 29 cm away. It is also to the side of and below the radiator, about 48 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0085_00_7", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 28.5}, {\"object\": \"radiator\", \"relation\": \"side_below\", \"distance_cm\": 47.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0085_00_8", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0085_00_9", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. lamp\nB. pillows\nC. alarm clock\nD. pillow", "choices": ["A. lamp", "B. pillows", "C. alarm clock", "D. pillow"], "answer": "C. alarm clock", "target": "telephone", "split": "train"}
{"id": "scene0085_00_10", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the nightstand. It is also near the alarm clock, about 22 cm away. It is further to the side of and below the pillows, about 31 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0085_00_11", "scene_id": "scene0085_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"alarm clock\", \"relation\": \"near\", \"distance_cm\": 21.5}, {\"object\": \"pillows\", \"relation\": \"side_below\", \"distance_cm\": 31.1}]}", "target": "telephone", "split": "train"}
{"id": "scene0092_00_0", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0092_00_1", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. computer tower\nB. board\nC. desk\nD. guitar", "choices": ["A. computer tower", "B. board", "C. desk", "D. guitar"], "answer": "B. board", "target": "backpack", "split": "train"}
{"id": "scene0092_00_2", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the board, about 35 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0092_00_3", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"mail\", \"references\": [{\"object\": \"board\", \"relation\": \"side_below\", \"distance_cm\": 35.0}]}", "target": "backpack", "split": "train"}
{"id": "scene0092_00_4", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "scene0092_00_5", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the guitar in this scene?\nA. monitor\nB. nightstand\nC. laundry hamper\nD. lamp", "choices": ["A. monitor", "B. nightstand", "C. laundry hamper", "D. lamp"], "answer": "B. nightstand", "target": "guitar", "split": "train"}
{"id": "scene0092_00_6", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the guitar in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The guitar is to the side of and above the nightstand, about 8 cm away. It is also to the side of and below the lamp, about 20 cm away.", "target": "guitar", "split": "train"}
{"id": "scene0092_00_7", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the guitar in structured JSON format.", "choices": null, "answer": "{\"target\": \"guitar\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"nightstand\", \"relation\": \"side_above\", \"distance_cm\": 8.1}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 20.5}]}", "target": "guitar", "split": "train"}
{"id": "scene0092_00_8", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "scene0092_00_9", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. board\nB. desk\nC. computer tower\nD. backpack", "choices": ["A. board", "B. desk", "C. computer tower", "D. backpack"], "answer": "C. computer tower", "target": "keyboard", "split": "train"}
{"id": "scene0092_00_10", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and above the computer tower, about 37 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0092_00_11", "scene_id": "scene0092_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 36.6}]}", "target": "keyboard", "split": "train"}
{"id": "scene0093_00_0", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "scene0093_00_1", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. table\nB. chair\nC. monitor\nD. keyboard", "choices": ["A. table", "B. chair", "C. monitor", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "scene0093_00_2", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 5 cm away. It is also to the side of and below the monitor, about 32 cm away. It is further to the side of and above the chair, about 38 cm away.", "target": "mouse", "split": "train"}
{"id": "scene0093_00_3", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 31.8}]}", "target": "mouse", "split": "train"}
{"id": "scene0093_00_4", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0093_00_5", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. chair\nB. monitor\nC. mouse\nD. keyboard", "choices": ["A. chair", "B. monitor", "C. mouse", "D. keyboard"], "answer": "D. keyboard", "target": "telephone", "split": "train"}
{"id": "scene0093_00_6", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is near the keyboard, about 29 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0093_00_7", "scene_id": "scene0093_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"monitor\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 29.3}]}", "target": "telephone", "split": "train"}
{"id": "scene0096_00_0", "scene_id": "scene0096_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "Yes", "target": "hat", "split": "train"}
{"id": "scene0096_00_1", "scene_id": "scene0096_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hat in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hat is to the side of and below the clothes, about 21 cm away. It is also to the side of and below the picture, about 21 cm away. It is further to the side of and below the mirror, about 49 cm away.", "target": "hat", "split": "train"}
{"id": "scene0096_00_2", "scene_id": "scene0096_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hat in structured JSON format.", "choices": null, "answer": "{\"target\": \"hat\", \"support_surface\": \"trunk\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_below\", \"distance_cm\": 21.1}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 21.2}]}", "target": "hat", "split": "train"}
{"id": "scene0096_00_3", "scene_id": "scene0096_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0096_00_4", "scene_id": "scene0096_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. radiator\nB. windowsill\nC. toilet\nD. jar", "choices": ["A. radiator", "B. windowsill", "C. toilet", "D. jar"], "answer": "B. windowsill", "target": "toilet paper", "split": "train"}
{"id": "scene0096_00_5", "scene_id": "scene0096_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the windowsill, about 7 cm away. It is also to the side of and above the radiator, about 27 cm away. It is further to the side of and above the toilet, about 45 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0096_00_6", "scene_id": "scene0096_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"windowsill\", \"references\": [{\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 7.4}, {\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 26.9}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0098_00_0", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0098_00_1", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. shelf\nB. tube\nC. cup\nD. box", "choices": ["A. shelf", "B. tube", "C. cup", "D. box"], "answer": "C. cup", "target": "bag", "split": "train"}
{"id": "scene0098_00_2", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the cup, about 9 cm away. It is also to the side of and below the tube, about 20 cm away. It is further to the side of and below the box, about 40 cm away.", "target": "bag", "split": "train"}
{"id": "scene0098_00_3", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"cup\", \"relation\": \"side_below\", \"distance_cm\": 9.3}, {\"object\": \"tube\", \"relation\": \"side_below\", \"distance_cm\": 19.6}]}", "target": "bag", "split": "train"}
{"id": "scene0098_00_4", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the shoe appear?", "choices": null, "answer": "No", "target": "shoe", "split": "train"}
{"id": "scene0098_00_5", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the computer tower, about 18 cm away. It is also to the side of and below the monitor, about 18 cm away. It is further to the side of and above the chair, about 51 cm away.", "target": "mouse", "split": "train"}
{"id": "scene0098_00_6", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 17.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.5}]}", "target": "mouse", "split": "train"}
{"id": "scene0098_00_7", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the coffee kettle appear?", "choices": null, "answer": "No", "target": "coffee kettle", "split": "train"}
{"id": "scene0098_00_8", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. cup\nB. box\nC. tube\nD. bag", "choices": ["A. cup", "B. box", "C. tube", "D. bag"], "answer": "B. box", "target": "water bottle", "split": "train"}
{"id": "scene0098_00_9", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the shelf. It is also to the side of and above the box, about 47 cm away. It is further to the side of and above the cup, about 69 cm away.", "target": "water bottle", "split": "train"}
{"id": "scene0098_00_10", "scene_id": "scene0098_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 46.7}, {\"object\": \"cup\", \"relation\": \"side_above\", \"distance_cm\": 68.5}]}", "target": "water bottle", "split": "train"}
{"id": "scene0102_00_0", "scene_id": "scene0102_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0102_00_1", "scene_id": "scene0102_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. chair\nB. monitor\nC. desk\nD. backpack", "choices": ["A. chair", "B. monitor", "C. desk", "D. backpack"], "answer": "A. chair", "target": "bag", "split": "train"}
{"id": "scene0105_00_0", "scene_id": "scene0105_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0105_00_1", "scene_id": "scene0105_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0105_00_2", "scene_id": "scene0105_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. trash can\nB. toilet\nC. shower walls\nD. shampoo", "choices": ["A. trash can", "B. toilet", "C. shower walls", "D. shampoo"], "answer": "A. trash can", "target": "toilet paper", "split": "train"}
{"id": "scene0105_00_3", "scene_id": "scene0105_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the trash can, about 14 cm away. It is also to the side of and above the toilet, about 22 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0105_00_4", "scene_id": "scene0105_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"trash can\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 13.7}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 22.2}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0107_00_0", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0107_00_1", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. lamp\nB. shoes\nC. suitcase\nD. nightstand", "choices": ["A. lamp", "B. shoes", "C. suitcase", "D. nightstand"], "answer": "C. suitcase", "target": "backpack", "split": "train"}
{"id": "scene0107_00_2", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the suitcase, about 7 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0107_00_3", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"suitcase\", \"references\": [{\"object\": \"suitcase\", \"relation\": \"side_below\", \"distance_cm\": 7.2}]}", "target": "backpack", "split": "train"}
{"id": "scene0107_00_4", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the candle appear?", "choices": null, "answer": "No", "target": "candle", "split": "train"}
{"id": "scene0107_00_5", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. shoes\nB. backpack\nC. nightstand\nD. lamp", "choices": ["A. shoes", "B. backpack", "C. nightstand", "D. lamp"], "answer": "D. lamp", "target": "telephone", "split": "train"}
{"id": "scene0107_00_6", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the nightstand. It is also to the side of and below the lamp, about 16 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0107_00_7", "scene_id": "scene0107_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 15.9}]}", "target": "telephone", "split": "train"}
{"id": "scene0108_00_0", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "scene0108_00_1", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. laptop\nB. chair\nC. water bottle\nD. computer tower", "choices": ["A. laptop", "B. chair", "C. water bottle", "D. computer tower"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0108_00_2", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 25 cm away. It is also to the side of and below the computer tower, about 35 cm away. It is further to the side of and below the laptop, about 38 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0108_00_3", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 24.9}, {\"object\": \"computer tower\", \"relation\": \"side_below\", \"distance_cm\": 34.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0108_00_4", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plate appear?", "choices": null, "answer": "No", "target": "plate", "split": "train"}
{"id": "scene0108_00_5", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. box\nB. whiteboard\nC. backpack\nD. chair", "choices": ["A. box", "B. whiteboard", "C. backpack", "D. chair"], "answer": "D. chair", "target": "keyboard", "split": "train"}
{"id": "scene0108_00_6", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also near the chair, about 25 cm away. It is further to the side of and above the whiteboard, about 51 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0108_00_7", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"near\", \"distance_cm\": 25.3}, {\"object\": \"whiteboard\", \"relation\": \"side_above\", \"distance_cm\": 51.4}]}", "target": "keyboard", "split": "train"}
{"id": "scene0108_00_8", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "train"}
{"id": "scene0108_00_9", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. backpack\nB. laptop\nC. box\nD. computer tower", "choices": ["A. backpack", "B. laptop", "C. box", "D. computer tower"], "answer": "B. laptop", "target": "water bottle", "split": "train"}
{"id": "scene0108_00_10", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the desk. It is also to the side of and above the laptop, about 12 cm away. It is further to the side of and above the computer tower, about 29 cm away.", "target": "water bottle", "split": "train"}
{"id": "scene0108_00_11", "scene_id": "scene0108_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 11.7}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 29.1}]}", "target": "water bottle", "split": "train"}
{"id": "scene0110_00_0", "scene_id": "scene0110_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0110_00_1", "scene_id": "scene0110_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. shelf\nC. desk\nD. chair", "choices": ["A. monitor", "B. shelf", "C. desk", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "train"}
{"id": "scene0110_00_2", "scene_id": "scene0110_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the desk. It is also to the side of and above the chair, about 39 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0110_00_3", "scene_id": "scene0110_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 39.0}]}", "target": "backpack", "split": "train"}
{"id": "scene0111_00_0", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0111_00_1", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bowl\nB. bag\nC. dish rack\nD. stool", "choices": ["A. bowl", "B. bag", "C. dish rack", "D. stool"], "answer": "D. stool", "target": "backpack", "split": "train"}
{"id": "scene0111_00_2", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the stool, about 54 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0111_00_3", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"stool\", \"relation\": \"side_below\", \"distance_cm\": 54.4}]}", "target": "backpack", "split": "train"}
{"id": "scene0111_00_4", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0111_00_5", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. table\nB. stool\nC. bowl\nD. shoes", "choices": ["A. table", "B. stool", "C. bowl", "D. shoes"], "answer": "A. table", "target": "bag", "split": "train"}
{"id": "scene0111_00_6", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the table, about 9 cm away. It is also to the side of and above the shoes, about 11 cm away. It is further to the side of and below the stool, about 48 cm away.", "target": "bag", "split": "train"}
{"id": "scene0111_00_7", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 9.1}, {\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 10.7}]}", "target": "bag", "split": "train"}
{"id": "scene0111_00_8", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "Yes", "target": "bowl", "split": "train"}
{"id": "scene0111_00_9", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bowl in this scene?\nA. grocery bag\nB. bag\nC. recycling bin\nD. shoes", "choices": ["A. grocery bag", "B. bag", "C. recycling bin", "D. shoes"], "answer": "C. recycling bin", "target": "bowl", "split": "train"}
{"id": "scene0111_00_10", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bowl in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bowl is on the table. It is also to the side of and above the recycling bin, about 32 cm away. It is further to the side of and above the grocery bag, about 46 cm away.", "target": "bowl", "split": "train"}
{"id": "scene0111_00_11", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bowl in structured JSON format.", "choices": null, "answer": "{\"target\": \"bowl\", \"support_surface\": \"table\", \"references\": [{\"object\": \"recycling bin\", \"relation\": \"side_above\", \"distance_cm\": 31.5}, {\"object\": \"grocery bag\", \"relation\": \"side_above\", \"distance_cm\": 45.7}]}", "target": "bowl", "split": "train"}
{"id": "scene0111_00_12", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the shoe appear?", "choices": null, "answer": "No", "target": "shoe", "split": "train"}
{"id": "scene0111_00_13", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scale in this scene?\nA. bowl\nB. tv stand\nC. bag\nD. backpack", "choices": ["A. bowl", "B. tv stand", "C. bag", "D. backpack"], "answer": "B. tv stand", "target": "scale", "split": "train"}
{"id": "scene0111_00_14", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is to the side of and below the tv stand, about 14 cm away.", "target": "scale", "split": "train"}
{"id": "scene0111_00_15", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"tv stand\", \"references\": [{\"object\": \"tv stand\", \"relation\": \"side_below\", \"distance_cm\": 13.6}]}", "target": "scale", "split": "train"}
{"id": "scene0111_00_16", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "train"}
{"id": "scene0111_00_17", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. tv\nB. bag\nC. backpack\nD. bowl", "choices": ["A. tv", "B. bag", "C. backpack", "D. bowl"], "answer": "A. tv", "target": "telephone", "split": "train"}
{"id": "scene0111_00_18", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the tv, about 55 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0111_00_19", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"tv\", \"references\": [{\"object\": \"tv\", \"relation\": \"side_above\", \"distance_cm\": 54.8}]}", "target": "telephone", "split": "train"}
{"id": "scene0111_00_20", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0111_00_21", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toaster in this scene?\nA. backpack\nB. sink\nC. bag\nD. dish rack", "choices": ["A. backpack", "B. sink", "C. bag", "D. dish rack"], "answer": "D. dish rack", "target": "toaster", "split": "train"}
{"id": "scene0111_00_22", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toaster in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster is next to the dish rack, about 6 cm away. It is also to the side of and above the sink, about 48 cm away.", "target": "toaster", "split": "train"}
{"id": "scene0111_00_23", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toaster in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster\", \"support_surface\": \"dish rack\", \"references\": [{\"object\": \"dish rack\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 48.4}]}", "target": "toaster", "split": "train"}
{"id": "scene0111_00_24", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "No", "target": "book", "split": "train"}
{"id": "scene0111_00_25", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "25", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. bag\nB. recycling bin\nC. food bag\nD. grocery bag", "choices": ["A. bag", "B. recycling bin", "C. food bag", "D. grocery bag"], "answer": "C. food bag", "target": "water bottle", "split": "train"}
{"id": "scene0111_00_26", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "26", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also near the food bag, about 18 cm away. It is further to the side of and above the recycling bin, about 46 cm away.", "target": "water bottle", "split": "train"}
{"id": "scene0111_00_27", "scene_id": "scene0111_00", "source_dataset": "scannet200", "local_sample_id": "27", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"food bag\", \"relation\": \"near\", \"distance_cm\": 17.5}, {\"object\": \"recycling bin\", \"relation\": \"side_above\", \"distance_cm\": 46.0}]}", "target": "water bottle", "split": "train"}
{"id": "scene0114_00_0", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "scene0114_00_1", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. monitor\nB. telephone\nC. desk\nD. file cabinet", "choices": ["A. monitor", "B. telephone", "C. desk", "D. file cabinet"], "answer": "B. telephone", "target": "cup", "split": "train"}
{"id": "scene0114_00_2", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the telephone, about 34 cm away. It is also to the side of and above the file cabinet, about 45 cm away.", "target": "cup", "split": "train"}
{"id": "scene0114_00_3", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"telephone\", \"relation\": \"side_below\", \"distance_cm\": 33.7}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 45.4}]}", "target": "cup", "split": "train"}
{"id": "scene0114_00_4", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "train"}
{"id": "scene0114_00_5", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. desk\nB. cup\nC. file cabinet\nD. monitor", "choices": ["A. desk", "B. cup", "C. file cabinet", "D. monitor"], "answer": "B. cup", "target": "telephone", "split": "train"}
{"id": "scene0114_00_6", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also to the side of and above the cup, about 34 cm away. It is further to the side of and below the monitor, about 35 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0114_00_7", "scene_id": "scene0114_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cup\", \"relation\": \"side_above\", \"distance_cm\": 33.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 34.8}]}", "target": "telephone", "split": "train"}
{"id": "scene0116_00_0", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "Yes", "target": "soap dish", "split": "train"}
{"id": "scene0116_00_1", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. shower walls\nB. towel\nC. picture\nD. bathtub", "choices": ["A. shower walls", "B. towel", "C. picture", "D. bathtub"], "answer": "D. bathtub", "target": "soap dish", "split": "train"}
{"id": "scene0116_00_2", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the bathtub, about 13 cm away. It is also to the side of and above the towel, about 53 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0116_00_3", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"bathtub\", \"references\": [{\"object\": \"bathtub\", \"relation\": \"side_above\", \"distance_cm\": 12.9}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 53.3}]}", "target": "soap dish", "split": "train"}
{"id": "scene0116_00_4", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "train"}
{"id": "scene0116_00_5", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. shower walls\nB. toilet\nC. bathtub\nD. picture", "choices": ["A. shower walls", "B. toilet", "C. bathtub", "D. picture"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0116_00_6", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 44 cm away. It is also below the picture, about 73 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0116_00_7", "scene_id": "scene0116_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 44.4}, {\"object\": \"picture\", \"relation\": \"below\", \"distance_cm\": 73.2}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0124_00_0", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "Yes", "target": "soap dispenser", "split": "train"}
{"id": "scene0124_00_1", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. light\nB. towel\nC. toilet\nD. soap", "choices": ["A. light", "B. towel", "C. toilet", "D. soap"], "answer": "D. soap", "target": "soap dispenser", "split": "train"}
{"id": "scene0124_00_2", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is next to the soap, about 5 cm away. It is also to the side of and below the light, about 44 cm away. It is further to the side of and above the toilet, about 57 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0124_00_3", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"soap\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"light\", \"relation\": \"side_below\", \"distance_cm\": 44.3}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0124_00_4", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plunger appear?", "choices": null, "answer": "No", "target": "plunger", "split": "train"}
{"id": "scene0124_00_5", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. light\nB. trash can\nC. bathroom cabinet\nD. toilet", "choices": ["A. light", "B. trash can", "C. bathroom cabinet", "D. toilet"], "answer": "D. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0124_00_6", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 13 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0124_00_7", "scene_id": "scene0124_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"trash can\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 12.9}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0126_00_0", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "scene0126_00_1", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the mouse, about 7 cm away. It is further next to the keyboard, about 7 cm away.", "target": "cup", "split": "train"}
{"id": "scene0126_00_2", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 7.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.3}]}", "target": "cup", "split": "train"}
{"id": "scene0126_00_3", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "scene0126_00_4", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the cup, about 7 cm away. It is further next to the mouse, about 8 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0126_00_5", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 7.3}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 7.5}]}", "target": "keyboard", "split": "train"}
{"id": "scene0126_00_6", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0126_00_7", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the cup, about 7 cm away. It is further next to the keyboard, about 8 cm away.", "target": "mouse", "split": "train"}
{"id": "scene0126_00_8", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 7.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.5}]}", "target": "mouse", "split": "train"}
{"id": "scene0126_00_9", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0126_00_10", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. file cabinet\nB. cup\nC. mouse\nD. computer tower", "choices": ["A. file cabinet", "B. cup", "C. mouse", "D. computer tower"], "answer": "A. file cabinet", "target": "telephone", "split": "train"}
{"id": "scene0126_00_11", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also to the side of and above the file cabinet, about 13 cm away. It is further near the cup, about 34 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0126_00_12", "scene_id": "scene0126_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 13.1}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 33.9}]}", "target": "telephone", "split": "train"}
{"id": "scene0130_00_0", "scene_id": "scene0130_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0131_00_0", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0131_00_1", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. computer tower\nB. office chair\nC. monitor\nD. board", "choices": ["A. computer tower", "B. office chair", "C. monitor", "D. board"], "answer": "A. computer tower", "target": "cup", "split": "train"}
{"id": "scene0131_00_2", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also above the computer tower, about 28 cm away. It is further to the side of and below the monitor, about 45 cm away.", "target": "cup", "split": "train"}
{"id": "scene0131_00_3", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"above\", \"distance_cm\": 27.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 45.2}]}", "target": "cup", "split": "train"}
{"id": "scene0131_00_4", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper towel roll appear?", "choices": null, "answer": "No", "target": "paper towel roll", "split": "train"}
{"id": "scene0131_00_5", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. cup\nC. computer tower\nD. board", "choices": ["A. chair", "B. cup", "C. computer tower", "D. board"], "answer": "C. computer tower", "target": "keyboard", "split": "train"}
{"id": "scene0131_00_6", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the computer tower, about 31 cm away. It is also to the side of and above the chair, about 45 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0131_00_7", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 31.4}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 44.8}]}", "target": "keyboard", "split": "train"}
{"id": "scene0131_00_8", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plate appear?", "choices": null, "answer": "Yes", "target": "plate", "split": "train"}
{"id": "scene0131_00_9", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plate in this scene?\nA. chair\nB. whiteboard\nC. computer tower\nD. board", "choices": ["A. chair", "B. whiteboard", "C. computer tower", "D. board"], "answer": "B. whiteboard", "target": "plate", "split": "train"}
{"id": "scene0131_00_10", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plate in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plate is to the side of and above the board, about 27 cm away.", "target": "plate", "split": "train"}
{"id": "scene0131_00_11", "scene_id": "scene0131_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plate in structured JSON format.", "choices": null, "answer": "{\"target\": \"plate\", \"support_surface\": \"mini fridge\", \"references\": [{\"object\": \"board\", \"relation\": \"side_above\", \"distance_cm\": 26.6}]}", "target": "plate", "split": "train"}
{"id": "scene0138_00_0", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0138_00_1", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. chair\nB. monitor\nC. cabinet\nD. computer tower", "choices": ["A. chair", "B. monitor", "C. cabinet", "D. computer tower"], "answer": "B. monitor", "target": "book", "split": "train"}
{"id": "scene0138_00_2", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the desk. It is also to the side of and below the monitor, about 30 cm away.", "target": "book", "split": "train"}
{"id": "scene0138_00_3", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 30.2}]}", "target": "book", "split": "train"}
{"id": "scene0138_00_4", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "scene0138_00_5", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"cabinet\", \"references\": []}", "target": "cup", "split": "train"}
{"id": "scene0138_00_6", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0138_00_7", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. computer tower\nB. book\nC. chair\nD. monitor", "choices": ["A. computer tower", "B. book", "C. chair", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0138_00_8", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and below the monitor, about 9 cm away. It is further to the side of and above the chair, about 23 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0138_00_9", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 22.6}]}", "target": "keyboard", "split": "train"}
{"id": "scene0138_00_10", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0138_00_11", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. computer tower\nB. monitor\nC. cabinet\nD. chair", "choices": ["A. computer tower", "B. monitor", "C. cabinet", "D. chair"], "answer": "D. chair", "target": "laptop", "split": "train"}
{"id": "scene0138_00_12", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also to the side of and above the chair, about 4 cm away. It is further to the side of and above the computer tower, about 23 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0138_00_13", "scene_id": "scene0138_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 4.2}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 23.0}]}", "target": "laptop", "split": "train"}
{"id": "scene0139_00_0", "scene_id": "scene0139_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0139_00_1", "scene_id": "scene0139_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. sink\nB. chair\nC. monitor\nD. paper towel dispenser", "choices": ["A. sink", "B. chair", "C. monitor", "D. paper towel dispenser"], "answer": "A. sink", "target": "soap dispenser", "split": "train"}
{"id": "scene0139_00_2", "scene_id": "scene0139_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the sink, about 38 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0139_00_3", "scene_id": "scene0139_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"paper towel dispenser\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 38.2}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0141_00_0", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0141_00_1", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. cup\nC. tube\nD. shoe", "choices": ["A. chair", "B. cup", "C. tube", "D. shoe"], "answer": "A. chair", "target": "backpack", "split": "train"}
{"id": "scene0141_00_2", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 23 cm away. It is also to the side of and below the cup, about 28 cm away. It is further to the side of and above the shoe, about 45 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0141_00_3", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 23.2}, {\"object\": \"cup\", \"relation\": \"side_below\", \"distance_cm\": 27.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0141_00_4", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0141_00_5", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. file cabinet\nB. book\nC. desk\nD. chair", "choices": ["A. file cabinet", "B. book", "C. desk", "D. chair"], "answer": "D. chair", "target": "ball", "split": "train"}
{"id": "scene0141_00_6", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is under the chair, about 3 cm away. It is also to the side of and below the desk, about 35 cm away. It is further to the side of and above the file cabinet, about 45 cm away.", "target": "ball", "split": "train"}
{"id": "scene0141_00_7", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"chair\", \"relation\": \"under\", \"distance_cm\": 3.0}, {\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 35.0}]}", "target": "ball", "split": "train"}
{"id": "scene0141_00_8", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "No", "target": "telephone", "split": "train"}
{"id": "scene0141_00_9", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. desk\nB. water bottle\nC. keyboard\nD. monitor", "choices": ["A. desk", "B. water bottle", "C. keyboard", "D. monitor"], "answer": "D. monitor", "target": "book", "split": "train"}
{"id": "scene0141_00_10", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the desk. It is also under the monitor, about 5 cm away. It is further next to the water bottle, about 12 cm away.", "target": "book", "split": "train"}
{"id": "scene0141_00_11", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.6}, {\"object\": \"water bottle\", \"relation\": \"next_to\", \"distance_cm\": 12.3}]}", "target": "book", "split": "train"}
{"id": "scene0141_00_12", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "Yes", "target": "guitar", "split": "train"}
{"id": "scene0141_00_13", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the guitar in this scene?\nA. backpack\nB. chair\nC. ball\nD. file cabinet", "choices": ["A. backpack", "B. chair", "C. ball", "D. file cabinet"], "answer": "B. chair", "target": "guitar", "split": "train"}
{"id": "scene0141_00_14", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the guitar in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The guitar is to the side of and below the chair, about 27 cm away. It is also near the file cabinet, about 58 cm away.", "target": "guitar", "split": "train"}
{"id": "scene0141_00_15", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the guitar in structured JSON format.", "choices": null, "answer": "{\"target\": \"guitar\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 27.1}, {\"object\": \"file cabinet\", \"relation\": \"near\", \"distance_cm\": 57.7}]}", "target": "guitar", "split": "train"}
{"id": "scene0141_00_16", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0141_00_17", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. mouse\nC. books\nD. book", "choices": ["A. monitor", "B. mouse", "C. books", "D. book"], "answer": "D. book", "target": "keyboard", "split": "train"}
{"id": "scene0141_00_18", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also near the book, about 19 cm away. It is further to the side of and below the books, about 23 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0141_00_19", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 19.3}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 23.0}]}", "target": "keyboard", "split": "train"}
{"id": "scene0141_00_20", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "No", "target": "hair dryer", "split": "train"}
{"id": "scene0141_00_21", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. desk\nC. monitor\nD. water bottle", "choices": ["A. keyboard", "B. desk", "C. monitor", "D. water bottle"], "answer": "B. desk", "target": "mouse", "split": "train"}
{"id": "scene0141_00_22", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the desk, about 8 cm away. It is also to the side of and above the water bottle, about 23 cm away. It is further to the side of and below the monitor, about 27 cm away.", "target": "mouse", "split": "train"}
{"id": "scene0141_00_23", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 7.9}, {\"object\": \"water bottle\", \"relation\": \"side_above\", \"distance_cm\": 22.6}]}", "target": "mouse", "split": "train"}
{"id": "scene0141_00_24", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the shoe appear?", "choices": null, "answer": "Yes", "target": "shoe", "split": "train"}
{"id": "scene0141_00_25", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "25", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the shoe in this scene?\nA. backpack\nB. chair\nC. file cabinet\nD. monitor", "choices": ["A. backpack", "B. chair", "C. file cabinet", "D. monitor"], "answer": "C. file cabinet", "target": "shoe", "split": "train"}
{"id": "scene0141_00_26", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "26", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the shoe in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The shoe is to the side of and below the file cabinet, about 11 cm away. It is also to the side of and below the chair, about 26 cm away. It is further to the side of and below the backpack, about 45 cm away.", "target": "shoe", "split": "train"}
{"id": "scene0141_00_27", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "27", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the shoe in structured JSON format.", "choices": null, "answer": "{\"target\": \"shoe\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"file cabinet\", \"relation\": \"side_below\", \"distance_cm\": 11.3}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 26.4}]}", "target": "shoe", "split": "train"}
{"id": "scene0141_00_28", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "28", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0141_00_29", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "29", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. mouse\nB. books\nC. file cabinet\nD. book", "choices": ["A. mouse", "B. books", "C. file cabinet", "D. book"], "answer": "C. file cabinet", "target": "water bottle", "split": "train"}
{"id": "scene0141_00_30", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "30", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the desk. It is also to the side of and above the file cabinet, about 9 cm away. It is further next to the book, about 12 cm away.", "target": "water bottle", "split": "train"}
{"id": "scene0141_00_31", "scene_id": "scene0141_00", "source_dataset": "scannet200", "local_sample_id": "31", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 8.5}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 12.3}]}", "target": "water bottle", "split": "train"}
{"id": "scene0146_00_0", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0146_00_1", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plate appear?", "choices": null, "answer": "No", "target": "plate", "split": "train"}
{"id": "scene0146_00_2", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. sink\nB. backpack\nC. mirror\nD. toilet", "choices": ["A. sink", "B. backpack", "C. mirror", "D. toilet"], "answer": "A. sink", "target": "soap dispenser", "split": "train"}
{"id": "scene0146_00_3", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the sink, about 21 cm away. It is also near the mirror, about 22 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0146_00_4", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"sink\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 20.5}, {\"object\": \"mirror\", \"relation\": \"near\", \"distance_cm\": 21.9}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0146_00_5", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the fire extinguisher appear?", "choices": null, "answer": "No", "target": "fire extinguisher", "split": "train"}
{"id": "scene0146_00_6", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. trash can\nB. toilet\nC. mirror\nD. backpack", "choices": ["A. trash can", "B. toilet", "C. mirror", "D. backpack"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0146_00_7", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 30 cm away. It is also near the trash can, about 70 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0146_00_8", "scene_id": "scene0146_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet paper dispenser\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 29.9}, {\"object\": \"trash can\", \"relation\": \"near\", \"distance_cm\": 70.2}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0147_00_0", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0147_00_1", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. hat\nB. tv\nC. picture\nD. dresser", "choices": ["A. hat", "B. tv", "C. picture", "D. dresser"], "answer": "A. hat", "target": "bag", "split": "train"}
{"id": "scene0147_00_2", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the hat, about 64 cm away.", "target": "bag", "split": "train"}
{"id": "scene0147_00_3", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"hat\", \"relation\": \"side_below\", \"distance_cm\": 64.5}]}", "target": "bag", "split": "train"}
{"id": "scene0147_00_4", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "scene0147_00_5", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hat in this scene?\nA. picture\nB. dresser\nC. bag\nD. tv", "choices": ["A. picture", "B. dresser", "C. bag", "D. tv"], "answer": "A. picture", "target": "hat", "split": "train"}
{"id": "scene0147_00_6", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hat in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hat is on the dresser. It is also to the side of and below the picture, about 15 cm away. It is further to the side of and below the tv, about 22 cm away.", "target": "hat", "split": "train"}
{"id": "scene0147_00_7", "scene_id": "scene0147_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hat in structured JSON format.", "choices": null, "answer": "{\"target\": \"hat\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 14.6}, {\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 22.4}]}", "target": "hat", "split": "train"}
{"id": "scene0149_00_0", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0149_00_1", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. chair\nB. kitchen cabinet\nC. table\nD. trash can", "choices": ["A. chair", "B. kitchen cabinet", "C. table", "D. trash can"], "answer": "D. trash can", "target": "bag", "split": "train"}
{"id": "scene0149_00_2", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the trash can, about 32 cm away. It is also to the side of and below the table, about 40 cm away. It is further to the side of and below the chair, about 43 cm away.", "target": "bag", "split": "train"}
{"id": "scene0149_00_3", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 32.0}, {\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 39.9}]}", "target": "bag", "split": "train"}
{"id": "scene0149_00_4", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the candle appear?", "choices": null, "answer": "No", "target": "candle", "split": "train"}
{"id": "scene0149_00_5", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper towel roll in this scene?\nA. bag\nB. kitchen cabinets\nC. water pitcher\nD. chair", "choices": ["A. bag", "B. kitchen cabinets", "C. water pitcher", "D. chair"], "answer": "C. water pitcher", "target": "paper towel roll", "split": "train"}
{"id": "scene0149_00_6", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper towel roll in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper towel roll is next to the water pitcher, about 10 cm away. It is also to the side of and below the kitchen cabinets, about 21 cm away.", "target": "paper towel roll", "split": "train"}
{"id": "scene0149_00_7", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"water pitcher\", \"references\": [{\"object\": \"water pitcher\", \"relation\": \"next_to\", \"distance_cm\": 9.8}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_below\", \"distance_cm\": 20.9}]}", "target": "paper towel roll", "split": "train"}
{"id": "scene0149_00_8", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cd case appear?", "choices": null, "answer": "No", "target": "cd case", "split": "train"}
{"id": "scene0149_00_9", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water pitcher in this scene?\nA. kitchen cabinets\nB. bag\nC. chair\nD. paper towel roll", "choices": ["A. kitchen cabinets", "B. bag", "C. chair", "D. paper towel roll"], "answer": "D. paper towel roll", "target": "water pitcher", "split": "train"}
{"id": "scene0149_00_10", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water pitcher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water pitcher is next to the paper towel roll, about 10 cm away. It is also to the side of and below the kitchen cabinets, about 23 cm away.", "target": "water pitcher", "split": "train"}
{"id": "scene0149_00_11", "scene_id": "scene0149_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water pitcher in structured JSON format.", "choices": null, "answer": "{\"target\": \"water pitcher\", \"support_surface\": \"paper towel roll\", \"references\": [{\"object\": \"paper towel roll\", \"relation\": \"next_to\", \"distance_cm\": 9.8}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_below\", \"distance_cm\": 23.0}]}", "target": "water pitcher", "split": "train"}
{"id": "scene0152_00_0", "scene_id": "scene0152_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0152_00_1", "scene_id": "scene0152_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. chair\nC. cabinet\nD. desk", "choices": ["A. monitor", "B. chair", "C. cabinet", "D. desk"], "answer": "C. cabinet", "target": "backpack", "split": "train"}
{"id": "scene0152_00_2", "scene_id": "scene0152_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the cabinet, about 6 cm away. It is also to the side of and below the chair, about 53 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0152_00_3", "scene_id": "scene0152_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 5.8}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 53.5}]}", "target": "backpack", "split": "train"}
{"id": "scene0153_00_0", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plunger appear?", "choices": null, "answer": "Yes", "target": "plunger", "split": "train"}
{"id": "scene0153_00_1", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plunger in this scene?\nA. bathroom cabinet\nB. toilet\nC. mirror\nD. toilet paper", "choices": ["A. bathroom cabinet", "B. toilet", "C. mirror", "D. toilet paper"], "answer": "B. toilet", "target": "plunger", "split": "train"}
{"id": "scene0153_00_2", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plunger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plunger is to the side of and below the toilet, about 20 cm away. It is also to the side of and below the toilet paper, about 72 cm away.", "target": "plunger", "split": "train"}
{"id": "scene0153_00_3", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plunger in structured JSON format.", "choices": null, "answer": "{\"target\": \"plunger\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 19.7}, {\"object\": \"toilet paper\", \"relation\": \"side_below\", \"distance_cm\": 72.1}]}", "target": "plunger", "split": "train"}
{"id": "scene0153_00_4", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0153_00_5", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scale in this scene?\nA. towel\nB. mirror\nC. bathroom cabinet\nD. plunger", "choices": ["A. towel", "B. mirror", "C. bathroom cabinet", "D. plunger"], "answer": "A. towel", "target": "scale", "split": "train"}
{"id": "scene0153_00_6", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is to the side of and below the towel, about 54 cm away.", "target": "scale", "split": "train"}
{"id": "scene0153_00_7", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"towel\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 54.5}]}", "target": "scale", "split": "train"}
{"id": "scene0153_00_8", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "Yes", "target": "soap dispenser", "split": "train"}
{"id": "scene0153_00_9", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. sink\nB. towel\nC. toilet paper\nD. mirror", "choices": ["A. sink", "B. towel", "C. toilet paper", "D. mirror"], "answer": "A. sink", "target": "soap dispenser", "split": "train"}
{"id": "scene0153_00_10", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the sink, about 6 cm away. It is also below the mirror, about 28 cm away. It is further to the side of and above the toilet paper, about 66 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0153_00_11", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 5.9}, {\"object\": \"mirror\", \"relation\": \"below\", \"distance_cm\": 28.0}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0153_00_12", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0153_00_13", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. sink\nB. toilet\nC. soap dispenser\nD. plunger", "choices": ["A. sink", "B. toilet", "C. soap dispenser", "D. plunger"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0153_00_14", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 15 cm away. It is also to the side of and below the sink, about 35 cm away. It is further to the side of and below the soap dispenser, about 66 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0153_00_15", "scene_id": "scene0153_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 15.4}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 34.7}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0155_00_0", "scene_id": "scene0155_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0155_00_1", "scene_id": "scene0155_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. keyboard\nB. desk\nC. chair\nD. monitor", "choices": ["A. keyboard", "B. desk", "C. chair", "D. monitor"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "scene0155_00_2", "scene_id": "scene0155_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the chair, about 36 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0155_00_3", "scene_id": "scene0155_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 36.4}]}", "target": "backpack", "split": "train"}
{"id": "scene0158_00_0", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "scene0158_00_1", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the fire extinguisher in this scene?\nA. dish rack\nB. trash can\nC. counter\nD. water pitcher", "choices": ["A. dish rack", "B. trash can", "C. counter", "D. water pitcher"], "answer": "D. water pitcher", "target": "fire extinguisher", "split": "train"}
{"id": "scene0158_00_2", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the fire extinguisher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The fire extinguisher is to the side of and above the water pitcher, about 30 cm away. It is also to the side of and above the trash can, about 36 cm away. It is further to the side of and above the dish rack, about 58 cm away.", "target": "fire extinguisher", "split": "train"}
{"id": "scene0158_00_3", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the fire extinguisher in structured JSON format.", "choices": null, "answer": "{\"target\": \"fire extinguisher\", \"support_surface\": \"dish rack\", \"references\": [{\"object\": \"water pitcher\", \"relation\": \"side_above\", \"distance_cm\": 29.5}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 35.9}]}", "target": "fire extinguisher", "split": "train"}
{"id": "scene0158_00_4", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "scene0158_00_5", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. fire extinguisher\nB. kitchen cabinet\nC. counter\nD. dish rack", "choices": ["A. fire extinguisher", "B. kitchen cabinet", "C. counter", "D. dish rack"], "answer": "B. kitchen cabinet", "target": "paper bag", "split": "train"}
{"id": "scene0158_00_6", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is under the kitchen cabinet, about 3 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0158_00_7", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen cabinet\", \"relation\": \"under\", \"distance_cm\": 3.5}]}", "target": "paper bag", "split": "train"}
{"id": "scene0158_00_8", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "Yes", "target": "soap dispenser", "split": "train"}
{"id": "scene0158_00_9", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. sink\nB. dish rack\nC. counter\nD. kitchen cabinets", "choices": ["A. sink", "B. dish rack", "C. counter", "D. kitchen cabinets"], "answer": "C. counter", "target": "soap dispenser", "split": "train"}
{"id": "scene0158_00_10", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the counter, about 21 cm away. It is also to the side of and above the sink, about 25 cm away. It is further to the side of and below the kitchen cabinets, about 26 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0158_00_11", "scene_id": "scene0158_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"counter\", \"relation\": \"side_above\", \"distance_cm\": 20.9}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 24.7}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0165_00_0", "scene_id": "scene0165_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0165_00_1", "scene_id": "scene0165_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. table\nB. armchair\nC. chair\nD. monitor", "choices": ["A. table", "B. armchair", "C. chair", "D. monitor"], "answer": "B. armchair", "target": "tissue box", "split": "train"}
{"id": "scene0165_00_2", "scene_id": "scene0165_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also to the side of and above the armchair, about 19 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0165_00_3", "scene_id": "scene0165_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"armchair\", \"relation\": \"side_above\", \"distance_cm\": 19.5}]}", "target": "tissue box", "split": "train"}
{"id": "scene0167_00_0", "scene_id": "scene0167_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0167_00_1", "scene_id": "scene0167_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the coffee table.", "target": "backpack", "split": "train"}
{"id": "scene0167_00_2", "scene_id": "scene0167_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"coffee table\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0168_00_0", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0168_00_1", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. box\nB. desk\nC. ball\nD. chair", "choices": ["A. box", "B. desk", "C. ball", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "train"}
{"id": "scene0168_00_2", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 33 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0168_00_3", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 33.3}]}", "target": "backpack", "split": "train"}
{"id": "scene0168_00_4", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0168_00_5", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. monitor\nB. desk\nC. backpack\nD. box", "choices": ["A. monitor", "B. desk", "C. backpack", "D. box"], "answer": "A. monitor", "target": "ball", "split": "train"}
{"id": "scene0168_00_6", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is to the side of and below the monitor, about 13 cm away.", "target": "ball", "split": "train"}
{"id": "scene0168_00_7", "scene_id": "scene0168_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.2}]}", "target": "ball", "split": "train"}
{"id": "scene0169_00_0", "scene_id": "scene0169_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0169_00_1", "scene_id": "scene0169_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. desk\nC. chair\nD. keyboard", "choices": ["A. monitor", "B. desk", "C. chair", "D. keyboard"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "scene0173_00_0", "scene_id": "scene0173_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plate appear?", "choices": null, "answer": "No", "target": "plate", "split": "train"}
{"id": "scene0173_00_1", "scene_id": "scene0173_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0174_00_0", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0174_00_1", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. keyboard\nC. monitor\nD. stuffed animal", "choices": ["A. desk", "B. keyboard", "C. monitor", "D. stuffed animal"], "answer": "D. stuffed animal", "target": "backpack", "split": "train"}
{"id": "scene0174_00_2", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the stuffed animal, about 34 cm away. It is also to the side of and below the keyboard, about 40 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0174_00_3", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"stuffed animal\", \"relation\": \"side_below\", \"distance_cm\": 34.1}, {\"object\": \"keyboard\", \"relation\": \"side_below\", \"distance_cm\": 40.0}]}", "target": "backpack", "split": "train"}
{"id": "scene0174_00_4", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0174_00_5", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. stuffed animal\nC. backpack\nD. desk", "choices": ["A. monitor", "B. stuffed animal", "C. backpack", "D. desk"], "answer": "B. stuffed animal", "target": "keyboard", "split": "train"}
{"id": "scene0174_00_6", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the stuffed animal, about 3 cm away. It is further to the side of and above the backpack, about 40 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0174_00_7", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"stuffed animal\", \"relation\": \"under\", \"distance_cm\": 3.2}, {\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 40.0}]}", "target": "keyboard", "split": "train"}
{"id": "scene0174_00_8", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "scene0174_00_9", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stuffed animal in this scene?\nA. desk\nB. keyboard\nC. monitor\nD. backpack", "choices": ["A. desk", "B. keyboard", "C. monitor", "D. backpack"], "answer": "B. keyboard", "target": "stuffed animal", "split": "train"}
{"id": "scene0174_00_10", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stuffed animal in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stuffed animal is on the desk. It is also next to the keyboard, about 3 cm away. It is further next to the monitor, about 9 cm away.", "target": "stuffed animal", "split": "train"}
{"id": "scene0174_00_11", "scene_id": "scene0174_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stuffed animal in structured JSON format.", "choices": null, "answer": "{\"target\": \"stuffed animal\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"monitor\", \"relation\": \"next_to\", \"distance_cm\": 9.4}]}", "target": "stuffed animal", "split": "train"}
{"id": "scene0175_00_0", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0175_00_1", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. towel\nB. toilet paper\nC. shorts\nD. toilet", "choices": ["A. towel", "B. toilet paper", "C. shorts", "D. toilet"], "answer": "C. shorts", "target": "soap dispenser", "split": "train"}
{"id": "scene0175_00_2", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is next to the shorts, about 9 cm away. It is also to the side of and above the towel, about 35 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0175_00_3", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"shorts\", \"references\": [{\"object\": \"shorts\", \"relation\": \"next_to\", \"distance_cm\": 9.3}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 35.1}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0175_00_4", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0175_00_5", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. soap dispenser\nB. toilet\nC. towel\nD. shorts", "choices": ["A. soap dispenser", "B. toilet", "C. towel", "D. shorts"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0175_00_6", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 25 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0175_00_7", "scene_id": "scene0175_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 25.2}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0177_00_0", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0177_00_1", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. computer tower\nC. chair\nD. couch", "choices": ["A. desk", "B. computer tower", "C. chair", "D. couch"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "scene0177_00_2", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the couch. It is also near the chair, about 42 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0177_00_3", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"couch\", \"references\": [{\"object\": \"chair\", \"relation\": \"near\", \"distance_cm\": 41.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0177_00_4", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dumbbell appear?", "choices": null, "answer": "Yes", "target": "dumbbell", "split": "train"}
{"id": "scene0177_00_5", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0177_00_6", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. telephone\nB. monitor\nC. computer tower\nD. file cabinet", "choices": ["A. telephone", "B. monitor", "C. computer tower", "D. file cabinet"], "answer": "B. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0177_00_7", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and below the monitor, about 5 cm away. It is further to the side of and above the file cabinet, about 19 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0177_00_8", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.3}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 19.3}]}", "target": "keyboard", "split": "train"}
{"id": "scene0177_00_9", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the fire extinguisher appear?", "choices": null, "answer": "No", "target": "fire extinguisher", "split": "train"}
{"id": "scene0177_00_10", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. monitor\nB. trash can\nC. lamp\nD. backpack", "choices": ["A. monitor", "B. trash can", "C. lamp", "D. backpack"], "answer": "C. lamp", "target": "laptop", "split": "train"}
{"id": "scene0177_00_11", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the shelf. It is also to the side of and below the lamp, about 30 cm away. It is further to the side of and below the monitor, about 39 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0177_00_12", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 30.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 38.6}]}", "target": "laptop", "split": "train"}
{"id": "scene0177_00_13", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "scene0177_00_14", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. file cabinet\nC. keyboard\nD. radiator", "choices": ["A. monitor", "B. file cabinet", "C. keyboard", "D. radiator"], "answer": "A. monitor", "target": "telephone", "split": "train"}
{"id": "scene0177_00_15", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also to the side of and below the monitor, about 12 cm away. It is further to the side of and above the file cabinet, about 18 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0177_00_16", "scene_id": "scene0177_00", "source_dataset": "scannet200", "local_sample_id": "16", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.1}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 17.9}]}", "target": "telephone", "split": "train"}
{"id": "scene0178_00_0", "scene_id": "scene0178_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0178_00_1", "scene_id": "scene0178_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bed\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0179_00_0", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0179_00_1", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. keyboard\nB. monitor\nC. cup\nD. telephone", "choices": ["A. keyboard", "B. monitor", "C. cup", "D. telephone"], "answer": "D. telephone", "target": "book", "split": "train"}
{"id": "scene0179_00_2", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the desk. It is also near the telephone, about 29 cm away. It is further to the side of and below the monitor, about 45 cm away.", "target": "book", "split": "train"}
{"id": "scene0179_00_3", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"telephone\", \"relation\": \"near\", \"distance_cm\": 29.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 44.5}]}", "target": "book", "split": "train"}
{"id": "scene0179_00_4", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0179_00_5", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. chair\nB. telephone\nC. keyboard\nD. monitor", "choices": ["A. chair", "B. telephone", "C. keyboard", "D. monitor"], "answer": "C. keyboard", "target": "cup", "split": "train"}
{"id": "scene0179_00_6", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the keyboard, about 6 cm away. It is further next to the telephone, about 11 cm away.", "target": "cup", "split": "train"}
{"id": "scene0179_00_7", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 10.5}]}", "target": "cup", "split": "train"}
{"id": "scene0179_00_8", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0179_00_9", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. cup\nB. keyboard\nC. book\nD. monitor", "choices": ["A. cup", "B. keyboard", "C. book", "D. monitor"], "answer": "A. cup", "target": "telephone", "split": "train"}
{"id": "scene0179_00_10", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also next to the cup, about 11 cm away. It is further to the side of and below the monitor, about 24 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0179_00_11", "scene_id": "scene0179_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 10.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.5}]}", "target": "telephone", "split": "train"}
{"id": "scene0183_00_0", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0183_00_1", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. cup\nB. couch\nC. laptop\nD. pillow", "choices": ["A. cup", "B. couch", "C. laptop", "D. pillow"], "answer": "A. cup", "target": "book", "split": "train"}
{"id": "scene0183_00_2", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is next to the cup, about 6 cm away. It is also next to the laptop, about 15 cm away. It is further to the side of and above the pillow, about 31 cm away.", "target": "book", "split": "train"}
{"id": "scene0183_00_3", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 14.8}]}", "target": "book", "split": "train"}
{"id": "scene0183_00_4", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0183_00_5", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. pillow\nB. book\nC. cup\nD. plate", "choices": ["A. pillow", "B. book", "C. cup", "D. plate"], "answer": "C. cup", "target": "laptop", "split": "train"}
{"id": "scene0183_00_6", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the coffee table. It is also next to the cup, about 7 cm away. It is further next to the book, about 15 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0183_00_7", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 6.6}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 14.8}]}", "target": "laptop", "split": "train"}
{"id": "scene0183_00_8", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plate appear?", "choices": null, "answer": "Yes", "target": "plate", "split": "train"}
{"id": "scene0183_00_9", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plate in this scene?\nA. cup\nB. laptop\nC. book\nD. coffee table", "choices": ["A. cup", "B. laptop", "C. book", "D. coffee table"], "answer": "B. laptop", "target": "plate", "split": "train"}
{"id": "scene0183_00_10", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plate in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plate is on the coffee table. It is also near the laptop, about 29 cm away. It is further near the cup, about 64 cm away.", "target": "plate", "split": "train"}
{"id": "scene0183_00_11", "scene_id": "scene0183_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plate in structured JSON format.", "choices": null, "answer": "{\"target\": \"plate\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 29.3}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 63.7}]}", "target": "plate", "split": "train"}
{"id": "scene0184_00_0", "scene_id": "scene0184_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the shoe appear?", "choices": null, "answer": "No", "target": "shoe", "split": "train"}
{"id": "scene0184_00_1", "scene_id": "scene0184_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the alarm clock in this scene?\nA. lamp\nB. pillow\nC. monitor\nD. nightstand", "choices": ["A. lamp", "B. pillow", "C. monitor", "D. nightstand"], "answer": "D. nightstand", "target": "alarm clock", "split": "train"}
{"id": "scene0184_00_2", "scene_id": "scene0184_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the alarm clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The alarm clock is to the side of and above the nightstand, about 7 cm away. It is also near the pillow, about 35 cm away.", "target": "alarm clock", "split": "train"}
{"id": "scene0184_00_3", "scene_id": "scene0184_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the alarm clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"alarm clock\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"nightstand\", \"relation\": \"side_above\", \"distance_cm\": 7.1}, {\"object\": \"pillow\", \"relation\": \"near\", \"distance_cm\": 35.2}]}", "target": "alarm clock", "split": "train"}
{"id": "scene0186_00_0", "scene_id": "scene0186_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0186_00_1", "scene_id": "scene0186_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. shelf\nB. desk\nC. monitor\nD. chair", "choices": ["A. shelf", "B. desk", "C. monitor", "D. chair"], "answer": "C. monitor", "target": "telephone", "split": "train"}
{"id": "scene0186_00_2", "scene_id": "scene0186_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also next to the monitor, about 5 cm away. It is further to the side of and below the shelf, about 25 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0186_00_3", "scene_id": "scene0186_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"next_to\", \"distance_cm\": 4.7}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 25.5}]}", "target": "telephone", "split": "train"}
{"id": "scene0187_00_0", "scene_id": "scene0187_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0187_00_1", "scene_id": "scene0187_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. table\nC. monitor\nD. desk", "choices": ["A. chair", "B. table", "C. monitor", "D. desk"], "answer": "B. table", "target": "backpack", "split": "train"}
{"id": "scene0187_00_2", "scene_id": "scene0187_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is under the table, about 3 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0187_00_3", "scene_id": "scene0187_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"under\", \"distance_cm\": 3.4}]}", "target": "backpack", "split": "train"}
{"id": "scene0194_00_0", "scene_id": "scene0194_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "train"}
{"id": "scene0194_00_1", "scene_id": "scene0194_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. person\nB. monitor\nC. chair\nD. desk", "choices": ["A. person", "B. monitor", "C. chair", "D. desk"], "answer": "A. person", "target": "backpack", "split": "train"}
{"id": "scene0194_00_2", "scene_id": "scene0194_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the person, about 31 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0194_00_3", "scene_id": "scene0194_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"person\", \"references\": [{\"object\": \"person\", \"relation\": \"side_below\", \"distance_cm\": 31.3}]}", "target": "backpack", "split": "train"}
{"id": "scene0195_00_0", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0195_00_1", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. counter\nB. towel\nC. sink\nD. laundry basket", "choices": ["A. counter", "B. towel", "C. sink", "D. laundry basket"], "answer": "A. counter", "target": "bag", "split": "train"}
{"id": "scene0195_00_2", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is under the counter, about 4 cm away. It is also to the side of and below the towel, about 16 cm away. It is further to the side of and above the laundry basket, about 23 cm away.", "target": "bag", "split": "train"}
{"id": "scene0195_00_3", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"counter\", \"relation\": \"under\", \"distance_cm\": 3.5}, {\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 16.2}]}", "target": "bag", "split": "train"}
{"id": "scene0195_00_4", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0195_00_5", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bag\nB. bathroom cabinet\nC. toilet\nD. trash can", "choices": ["A. bag", "B. bathroom cabinet", "C. toilet", "D. trash can"], "answer": "C. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0195_00_6", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 17 cm away. It is also to the side of and above the trash can, about 73 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0195_00_7", "scene_id": "scene0195_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 17.1}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 73.0}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0197_00_0", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the fire extinguisher appear?", "choices": null, "answer": "No", "target": "fire extinguisher", "split": "train"}
{"id": "scene0197_00_1", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. water cooler\nB. paper towel dispenser\nC. bowl\nD. tupperware", "choices": ["A. water cooler", "B. paper towel dispenser", "C. bowl", "D. tupperware"], "answer": "C. bowl", "target": "bottle", "split": "train"}
{"id": "scene0197_00_2", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the counter. It is also next to the bowl, about 3 cm away. It is further near the tupperware, about 21 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0197_00_3", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"bowl\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"tupperware\", \"relation\": \"near\", \"distance_cm\": 21.5}]}", "target": "bottle", "split": "train"}
{"id": "scene0197_00_4", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "Yes", "target": "bowl", "split": "train"}
{"id": "scene0197_00_5", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bowl in this scene?\nA. cabinet\nB. bottle\nC. water cooler\nD. paper towel dispenser", "choices": ["A. cabinet", "B. bottle", "C. water cooler", "D. paper towel dispenser"], "answer": "B. bottle", "target": "bowl", "split": "train"}
{"id": "scene0197_00_6", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bowl in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bowl is on the counter. It is also next to the bottle, about 3 cm away. It is further to the side of and below the water cooler, about 17 cm away.", "target": "bowl", "split": "train"}
{"id": "scene0197_00_7", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bowl in structured JSON format.", "choices": null, "answer": "{\"target\": \"bowl\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"water cooler\", \"relation\": \"side_below\", \"distance_cm\": 16.6}]}", "target": "bowl", "split": "train"}
{"id": "scene0197_00_8", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0197_00_9", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tray in this scene?\nA. printer\nB. water cooler\nC. shelf\nD. stapler", "choices": ["A. printer", "B. water cooler", "C. shelf", "D. stapler"], "answer": "B. water cooler", "target": "tray", "split": "train"}
{"id": "scene0197_00_10", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tray in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tray is on the counter. It is also to the side of and below the water cooler, about 8 cm away. It is further to the side of and above the stapler, about 15 cm away.", "target": "tray", "split": "train"}
{"id": "scene0197_00_11", "scene_id": "scene0197_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tray in structured JSON format.", "choices": null, "answer": "{\"target\": \"tray\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"water cooler\", \"relation\": \"side_below\", \"distance_cm\": 7.8}, {\"object\": \"stapler\", \"relation\": \"side_above\", \"distance_cm\": 14.7}]}", "target": "tray", "split": "train"}
{"id": "scene0198_00_0", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0198_00_1", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cup\nB. coffee maker\nC. kitchen counter\nD. dish rack", "choices": ["A. cup", "B. coffee maker", "C. kitchen counter", "D. dish rack"], "answer": "C. kitchen counter", "target": "backpack", "split": "train"}
{"id": "scene0198_00_2", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the kitchen counter, about 38 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0198_00_3", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_below\", \"distance_cm\": 37.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0198_00_4", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0198_00_5", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. kitchen cabinets\nB. paper towel dispenser\nC. coffee maker\nD. sink", "choices": ["A. kitchen cabinets", "B. paper towel dispenser", "C. coffee maker", "D. sink"], "answer": "C. coffee maker", "target": "cup", "split": "train"}
{"id": "scene0198_00_6", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the coffee maker, about 23 cm away. It is also to the side of and above the sink, about 36 cm away. It is further to the side of and above the kitchen cabinets, about 50 cm away.", "target": "cup", "split": "train"}
{"id": "scene0198_00_7", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"dish rack\", \"references\": [{\"object\": \"coffee maker\", \"relation\": \"side_below\", \"distance_cm\": 22.7}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 35.9}]}", "target": "cup", "split": "train"}
{"id": "scene0198_00_8", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0198_00_9", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper towel roll in this scene?\nA. kitchen cabinets\nB. cup\nC. microwave\nD. coffee maker", "choices": ["A. kitchen cabinets", "B. cup", "C. microwave", "D. coffee maker"], "answer": "A. kitchen cabinets", "target": "paper towel roll", "split": "train"}
{"id": "scene0198_00_10", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper towel roll in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper towel roll is to the side of and below the kitchen cabinets, about 35 cm away. It is also to the side of and below the microwave, about 38 cm away. It is further to the side of and below the coffee maker, about 49 cm away.", "target": "paper towel roll", "split": "train"}
{"id": "scene0198_00_11", "scene_id": "scene0198_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen cabinets\", \"relation\": \"side_below\", \"distance_cm\": 34.9}, {\"object\": \"microwave\", \"relation\": \"side_below\", \"distance_cm\": 37.8}]}", "target": "paper towel roll", "split": "train"}
{"id": "scene0201_00_0", "scene_id": "scene0201_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "scene0201_00_1", "scene_id": "scene0201_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. dining table\nC. chair\nD. desk", "choices": ["A. monitor", "B. dining table", "C. chair", "D. desk"], "answer": "B. dining table", "target": "telephone", "split": "train"}
{"id": "scene0201_00_2", "scene_id": "scene0201_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the dining table, about 38 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0201_00_3", "scene_id": "scene0201_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"dining table\", \"references\": [{\"object\": \"dining table\", \"relation\": \"side_above\", \"distance_cm\": 38.1}]}", "target": "telephone", "split": "train"}
{"id": "scene0205_00_0", "scene_id": "scene0205_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plunger appear?", "choices": null, "answer": "No", "target": "plunger", "split": "train"}
{"id": "scene0206_00_0", "scene_id": "scene0206_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the broom appear?", "choices": null, "answer": "No", "target": "broom", "split": "train"}
{"id": "scene0206_00_1", "scene_id": "scene0206_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. tripod\nB. stand\nC. chair\nD. monitor", "choices": ["A. tripod", "B. stand", "C. chair", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0206_00_2", "scene_id": "scene0206_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the monitor, about 7 cm away. It is further to the side of and below the tripod, about 29 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0206_00_3", "scene_id": "scene0206_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.6}, {\"object\": \"tripod\", \"relation\": \"side_below\", \"distance_cm\": 29.1}]}", "target": "keyboard", "split": "train"}
{"id": "scene0209_00_0", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0209_00_1", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. bed\nB. cabinet\nC. clothes\nD. shoes", "choices": ["A. bed", "B. cabinet", "C. clothes", "D. shoes"], "answer": "D. shoes", "target": "bag", "split": "train"}
{"id": "scene0209_00_2", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the shoes, about 7 cm away.", "target": "bag", "split": "train"}
{"id": "scene0209_00_3", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"suitcase\", \"references\": [{\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 7.5}]}", "target": "bag", "split": "train"}
{"id": "scene0209_00_4", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0209_00_5", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. bed\nB. clothes\nC. cabinet\nD. bag", "choices": ["A. bed", "B. clothes", "C. cabinet", "D. bag"], "answer": "B. clothes", "target": "laptop", "split": "train"}
{"id": "scene0209_00_6", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the bed. It is also to the side of and above the clothes, about 39 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0209_00_7", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_above\", \"distance_cm\": 38.9}]}", "target": "laptop", "split": "train"}
{"id": "scene0209_00_8", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0209_00_9", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. cabinet\nB. bed\nC. bag\nD. shopping bag", "choices": ["A. cabinet", "B. bed", "C. bag", "D. shopping bag"], "answer": "D. shopping bag", "target": "paper bag", "split": "train"}
{"id": "scene0209_00_10", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is to the side of and below the shopping bag, about 39 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0209_00_11", "scene_id": "scene0209_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"shopping bag\", \"relation\": \"side_below\", \"distance_cm\": 38.8}]}", "target": "paper bag", "split": "train"}
{"id": "scene0210_00_0", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "Yes", "target": "clock", "split": "train"}
{"id": "scene0210_00_1", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the clock in this scene?\nA. light switch\nB. cabinet\nC. fire alarm\nD. shelf", "choices": ["A. light switch", "B. cabinet", "C. fire alarm", "D. shelf"], "answer": "C. fire alarm", "target": "clock", "split": "train"}
{"id": "scene0210_00_2", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The clock is to the side of and below the fire alarm, about 32 cm away. It is also to the side of and above the light switch, about 44 cm away. It is further near the shelf, about 74 cm away.", "target": "clock", "split": "train"}
{"id": "scene0210_00_3", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"clock\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"fire alarm\", \"relation\": \"side_below\", \"distance_cm\": 31.5}, {\"object\": \"light switch\", \"relation\": \"side_above\", \"distance_cm\": 43.5}]}", "target": "clock", "split": "train"}
{"id": "scene0210_00_4", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0210_00_5", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper cutter in this scene?\nA. cabinet\nB. sink\nC. microwave\nD. dolly", "choices": ["A. cabinet", "B. sink", "C. microwave", "D. dolly"], "answer": "C. microwave", "target": "paper cutter", "split": "train"}
{"id": "scene0210_00_6", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper cutter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper cutter is under the microwave, about 4 cm away. It is also to the side of and above the cabinet, about 10 cm away. It is further to the side of and above the sink, about 19 cm away.", "target": "paper cutter", "split": "train"}
{"id": "scene0210_00_7", "scene_id": "scene0210_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper cutter in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper cutter\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"microwave\", \"relation\": \"under\", \"distance_cm\": 4.4}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 10.1}]}", "target": "paper cutter", "split": "train"}
{"id": "scene0214_00_0", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0214_00_1", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. dishwasher\nB. kitchen cabinets\nC. microwave\nD. sink", "choices": ["A. dishwasher", "B. kitchen cabinets", "C. microwave", "D. sink"], "answer": "D. sink", "target": "bottle", "split": "train"}
{"id": "scene0214_00_2", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the counter. It is also to the side of and above the sink, about 12 cm away. It is further to the side of and above the kitchen cabinets, about 27 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0214_00_3", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 11.9}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 27.0}]}", "target": "bottle", "split": "train"}
{"id": "scene0214_00_4", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0214_00_5", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bottle\nB. toilet\nC. shower\nD. shelf", "choices": ["A. bottle", "B. toilet", "C. shower", "D. shelf"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0214_00_6", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 28 cm away. It is also to the side of and below the shelf, about 41 cm away. It is further to the side of and below the shower, about 79 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0214_00_7", "scene_id": "scene0214_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 27.8}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 40.7}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0221_00_0", "scene_id": "scene0221_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0221_00_1", "scene_id": "scene0221_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. table\nB. monitor\nC. trash can\nD. chair", "choices": ["A. table", "B. monitor", "C. trash can", "D. chair"], "answer": "C. trash can", "target": "bag", "split": "train"}
{"id": "scene0221_00_2", "scene_id": "scene0221_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is on the table. It is also to the side of and above the trash can, about 36 cm away.", "target": "bag", "split": "train"}
{"id": "scene0221_00_3", "scene_id": "scene0221_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"table\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 35.7}]}", "target": "bag", "split": "train"}
{"id": "scene0223_00_0", "scene_id": "scene0223_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0223_00_1", "scene_id": "scene0223_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. chair\nB. monitor\nC. desk\nD. wardrobe closet", "choices": ["A. chair", "B. monitor", "C. desk", "D. wardrobe closet"], "answer": "D. wardrobe closet", "target": "telephone", "split": "train"}
{"id": "scene0223_00_2", "scene_id": "scene0223_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the wardrobe closet, about 9 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0223_00_3", "scene_id": "scene0223_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"wardrobe closet\", \"references\": [{\"object\": \"wardrobe closet\", \"relation\": \"side_above\", \"distance_cm\": 9.0}]}", "target": "telephone", "split": "train"}
{"id": "scene0227_00_0", "scene_id": "scene0227_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "scene0227_00_1", "scene_id": "scene0227_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. table\nC. desk\nD. chair", "choices": ["A. monitor", "B. table", "C. desk", "D. chair"], "answer": "D. chair", "target": "telephone", "split": "train"}
{"id": "scene0227_00_2", "scene_id": "scene0227_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the table. It is also near the chair, about 38 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0227_00_3", "scene_id": "scene0227_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"near\", \"distance_cm\": 38.4}]}", "target": "telephone", "split": "train"}
{"id": "scene0229_00_0", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0229_00_1", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. lamp\nB. books\nC. box\nD. storage bin", "choices": ["A. lamp", "B. books", "C. box", "D. storage bin"], "answer": "B. books", "target": "book", "split": "train"}
{"id": "scene0229_00_2", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is to the side of and below the books, about 9 cm away. It is also to the side of and below the storage bin, about 40 cm away.", "target": "book", "split": "train"}
{"id": "scene0229_00_3", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 8.7}, {\"object\": \"storage bin\", \"relation\": \"side_below\", \"distance_cm\": 40.0}]}", "target": "book", "split": "train"}
{"id": "scene0229_00_4", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0229_00_5", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the shoe in this scene?\nA. sewing machine\nB. lamp\nC. box\nD. table", "choices": ["A. sewing machine", "B. lamp", "C. box", "D. table"], "answer": "D. table", "target": "shoe", "split": "train"}
{"id": "scene0229_00_6", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the shoe in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The shoe is under the table, about 3 cm away. It is also below the sewing machine, about 65 cm away. It is further to the side of and below the lamp, about 68 cm away.", "target": "shoe", "split": "train"}
{"id": "scene0229_00_7", "scene_id": "scene0229_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the shoe in structured JSON format.", "choices": null, "answer": "{\"target\": \"shoe\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"under\", \"distance_cm\": 3.3}, {\"object\": \"sewing machine\", \"relation\": \"below\", \"distance_cm\": 65.1}]}", "target": "shoe", "split": "train"}
{"id": "scene0230_00_0", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plate appear?", "choices": null, "answer": "No", "target": "plate", "split": "train"}
{"id": "scene0230_00_1", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. chair\nB. trash can\nC. monitor\nD. bag", "choices": ["A. chair", "B. trash can", "C. monitor", "D. bag"], "answer": "D. bag", "target": "cup", "split": "train"}
{"id": "scene0230_00_2", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also under the bag, about 4 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "cup", "split": "train"}
{"id": "scene0230_00_3", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"bag\", \"relation\": \"under\", \"distance_cm\": 4.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.7}]}", "target": "cup", "split": "train"}
{"id": "scene0230_00_4", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "scene0230_00_5", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. trash can\nC. chair\nD. bag", "choices": ["A. monitor", "B. trash can", "C. chair", "D. bag"], "answer": "D. bag", "target": "keyboard", "split": "train"}
{"id": "scene0230_00_6", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the bag, about 5 cm away. It is further to the side of and below the monitor, about 17 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0230_00_7", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"bag\", \"relation\": \"under\", \"distance_cm\": 4.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 16.7}]}", "target": "keyboard", "split": "train"}
{"id": "scene0230_00_8", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0230_00_9", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. cup\nB. chair\nC. desk\nD. bag", "choices": ["A. cup", "B. chair", "C. desk", "D. bag"], "answer": "D. bag", "target": "laptop", "split": "train"}
{"id": "scene0230_00_10", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is to the side of and above the bag, about 49 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0230_00_11", "scene_id": "scene0230_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 49.3}]}", "target": "laptop", "split": "train"}
{"id": "scene0231_00_0", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0231_00_1", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. trash can\nB. armchair\nC. book\nD. bag", "choices": ["A. trash can", "B. armchair", "C. book", "D. bag"], "answer": "A. trash can", "target": "backpack", "split": "train"}
{"id": "scene0231_00_2", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the trash can, about 20 cm away. It is also to the side of and below the book, about 47 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0231_00_3", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 19.7}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 47.0}]}", "target": "backpack", "split": "train"}
{"id": "scene0231_00_4", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0231_00_5", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. lamp base\nB. backpack\nC. pillow\nD. armchair", "choices": ["A. lamp base", "B. backpack", "C. pillow", "D. armchair"], "answer": "A. lamp base", "target": "bag", "split": "train"}
{"id": "scene0231_00_6", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the lamp base, about 35 cm away. It is also to the side of and below the pillow, about 39 cm away.", "target": "bag", "split": "train"}
{"id": "scene0231_00_7", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"lamp base\", \"relation\": \"side_above\", \"distance_cm\": 35.0}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 38.7}]}", "target": "bag", "split": "train"}
{"id": "scene0231_00_8", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0231_00_9", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the clock in this scene?\nA. bag\nB. backpack\nC. rice cooker\nD. armchair", "choices": ["A. bag", "B. backpack", "C. rice cooker", "D. armchair"], "answer": "C. rice cooker", "target": "clock", "split": "train"}
{"id": "scene0231_00_10", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The clock is to the side of and above the rice cooker, about 47 cm away.", "target": "clock", "split": "train"}
{"id": "scene0231_00_11", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"clock\", \"support_surface\": \"rice cooker\", \"references\": [{\"object\": \"rice cooker\", \"relation\": \"side_above\", \"distance_cm\": 46.6}]}", "target": "clock", "split": "train"}
{"id": "scene0231_00_12", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0231_00_13", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. sink\nB. armchair\nC. box\nD. banana holder", "choices": ["A. sink", "B. armchair", "C. box", "D. banana holder"], "answer": "A. sink", "target": "tissue box", "split": "train"}
{"id": "scene0231_00_14", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and above the sink, about 22 cm away. It is also near the banana holder, about 50 cm away. It is further to the side of and above the box, about 50 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0231_00_15", "scene_id": "scene0231_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"box\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 21.9}, {\"object\": \"banana holder\", \"relation\": \"near\", \"distance_cm\": 49.5}]}", "target": "tissue box", "split": "train"}
{"id": "scene0232_00_0", "scene_id": "scene0232_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "No", "target": "bag", "split": "train"}
{"id": "scene0232_00_1", "scene_id": "scene0232_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper towel roll in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper towel roll is to the side of and above the clothes, about 27 cm away. It is also to the side of and above the clothes dryer, about 27 cm away.", "target": "paper towel roll", "split": "train"}
{"id": "scene0232_00_2", "scene_id": "scene0232_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_above\", \"distance_cm\": 27.1}, {\"object\": \"clothes dryer\", \"relation\": \"side_above\", \"distance_cm\": 27.3}]}", "target": "paper towel roll", "split": "train"}
{"id": "scene0234_00_0", "scene_id": "scene0234_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0234_00_1", "scene_id": "scene0234_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. lamp\nB. couch cushions\nC. table\nD. couch", "choices": ["A. lamp", "B. couch cushions", "C. table", "D. couch"], "answer": "A. lamp", "target": "book", "split": "train"}
{"id": "scene0234_00_2", "scene_id": "scene0234_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the table. It is also to the side of and below the lamp, about 13 cm away.", "target": "book", "split": "train"}
{"id": "scene0234_00_3", "scene_id": "scene0234_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"table\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 13.4}]}", "target": "book", "split": "train"}
{"id": "scene0235_00_0", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0235_00_1", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. lamp\nB. chair\nC. mirror\nD. backpack", "choices": ["A. lamp", "B. chair", "C. mirror", "D. backpack"], "answer": "C. mirror", "target": "laptop", "split": "train"}
{"id": "scene0235_00_2", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the table. It is also to the side of and below the mirror, about 17 cm away. It is further to the side of and above the chair, about 20 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0235_00_3", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 16.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 19.6}]}", "target": "laptop", "split": "train"}
{"id": "scene0235_00_4", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "train"}
{"id": "scene0235_00_5", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. sofa chair\nB. backpack\nC. lamp\nD. pillow", "choices": ["A. sofa chair", "B. backpack", "C. lamp", "D. pillow"], "answer": "A. sofa chair", "target": "telephone", "split": "train"}
{"id": "scene0235_00_6", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the nightstand. It is also to the side of and above the sofa chair, about 3 cm away. It is further to the side of and below the pillow, about 46 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0235_00_7", "scene_id": "scene0235_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"sofa chair\", \"relation\": \"side_above\", \"distance_cm\": 3.1}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 45.6}]}", "target": "telephone", "split": "train"}
{"id": "scene0238_00_0", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "train"}
{"id": "scene0238_00_1", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. computer tower\nB. desk\nC. monitor\nD. laptop", "choices": ["A. computer tower", "B. desk", "C. monitor", "D. laptop"], "answer": "C. monitor", "target": "cup", "split": "train"}
{"id": "scene0238_00_2", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also to the side of and below the monitor, about 24 cm away. It is further near the laptop, about 29 cm away.", "target": "cup", "split": "train"}
{"id": "scene0238_00_3", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.5}, {\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 28.6}]}", "target": "cup", "split": "train"}
{"id": "scene0238_00_4", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "scene0238_00_5", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. lamp\nB. file cabinet\nC. chair\nD. card", "choices": ["A. lamp", "B. file cabinet", "C. chair", "D. card"], "answer": "C. chair", "target": "keyboard", "split": "train"}
{"id": "scene0238_00_6", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and above the chair, about 24 cm away. It is further to the side of and below the card, about 47 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0238_00_7", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 24.3}, {\"object\": \"card\", \"relation\": \"side_below\", \"distance_cm\": 47.2}]}", "target": "keyboard", "split": "train"}
{"id": "scene0238_00_8", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0238_00_9", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. cup\nB. person\nC. computer tower\nD. file cabinet", "choices": ["A. cup", "B. person", "C. computer tower", "D. file cabinet"], "answer": "B. person", "target": "laptop", "split": "train"}
{"id": "scene0238_00_10", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also to the side of and above the person, about 21 cm away. It is further near the cup, about 29 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0238_00_11", "scene_id": "scene0238_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"person\", \"relation\": \"side_above\", \"distance_cm\": 20.7}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 28.6}]}", "target": "laptop", "split": "train"}
{"id": "scene0239_00_0", "scene_id": "scene0239_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0240_00_0", "scene_id": "scene0240_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0240_00_1", "scene_id": "scene0240_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the clock in this scene?\nA. monitor\nB. dustpan\nC. box\nD. shelf", "choices": ["A. monitor", "B. dustpan", "C. box", "D. shelf"], "answer": "C. box", "target": "clock", "split": "train"}
{"id": "scene0240_00_2", "scene_id": "scene0240_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The clock is to the side of and above the box, about 30 cm away. It is also to the side of and above the shelf, about 49 cm away.", "target": "clock", "split": "train"}
{"id": "scene0240_00_3", "scene_id": "scene0240_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"clock\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 30.1}, {\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 48.9}]}", "target": "clock", "split": "train"}
{"id": "scene0240_00_4", "scene_id": "scene0240_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dustpan appear?", "choices": null, "answer": "Yes", "target": "dustpan", "split": "train"}
{"id": "scene0241_00_0", "scene_id": "scene0241_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0241_00_1", "scene_id": "scene0241_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stuffed animal in this scene?\nA. bookshelf\nB. books\nC. monitor\nD. chair", "choices": ["A. bookshelf", "B. books", "C. monitor", "D. chair"], "answer": "B. books", "target": "stuffed animal", "split": "train"}
{"id": "scene0241_00_2", "scene_id": "scene0241_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stuffed animal in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stuffed animal is on the bookshelf. It is also above the books, about 12 cm away.", "target": "stuffed animal", "split": "train"}
{"id": "scene0241_00_3", "scene_id": "scene0241_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stuffed animal in structured JSON format.", "choices": null, "answer": "{\"target\": \"stuffed animal\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"books\", \"relation\": \"above\", \"distance_cm\": 11.5}]}", "target": "stuffed animal", "split": "train"}
{"id": "scene0242_00_0", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0242_00_1", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. fire extinguisher\nC. chair\nD. desk", "choices": ["A. monitor", "B. fire extinguisher", "C. chair", "D. desk"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "scene0242_00_2", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 45 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0242_00_3", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 44.5}]}", "target": "backpack", "split": "train"}
{"id": "scene0242_00_4", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0242_00_5", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the fire extinguisher in this scene?\nA. chair\nB. monitor\nC. backpack\nD. desk", "choices": ["A. chair", "B. monitor", "C. backpack", "D. desk"], "answer": "A. chair", "target": "fire extinguisher", "split": "train"}
{"id": "scene0242_00_6", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the fire extinguisher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The fire extinguisher is to the side of and above the chair, about 5 cm away.", "target": "fire extinguisher", "split": "train"}
{"id": "scene0242_00_7", "scene_id": "scene0242_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the fire extinguisher in structured JSON format.", "choices": null, "answer": "{\"target\": \"fire extinguisher\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 4.7}]}", "target": "fire extinguisher", "split": "train"}
{"id": "scene0243_00_0", "scene_id": "scene0243_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "train"}
{"id": "scene0243_00_1", "scene_id": "scene0243_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. clothes\nC. laundry basket\nD. shoes", "choices": ["A. bag", "B. clothes", "C. laundry basket", "D. shoes"], "answer": "B. clothes", "target": "backpack", "split": "train"}
{"id": "scene0243_00_2", "scene_id": "scene0243_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the clothes, about 26 cm away. It is also near the laundry basket, about 56 cm away. It is further to the side of and above the shoes, about 60 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0243_00_3", "scene_id": "scene0243_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_below\", \"distance_cm\": 26.5}, {\"object\": \"laundry basket\", \"relation\": \"near\", \"distance_cm\": 55.6}]}", "target": "backpack", "split": "train"}
{"id": "scene0250_00_0", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "Yes", "target": "hair dryer", "split": "train"}
{"id": "scene0250_00_1", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hair dryer in this scene?\nA. plant\nB. faucet\nC. trash can\nD. bottle", "choices": ["A. plant", "B. faucet", "C. trash can", "D. bottle"], "answer": "A. plant", "target": "hair dryer", "split": "train"}
{"id": "scene0250_00_2", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hair dryer in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hair dryer is under the plant, about 5 cm away. It is also to the side of and above the faucet, about 39 cm away. It is further to the side of and above the trash can, about 54 cm away.", "target": "hair dryer", "split": "train"}
{"id": "scene0250_00_3", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hair dryer in structured JSON format.", "choices": null, "answer": "{\"target\": \"hair dryer\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"plant\", \"relation\": \"under\", \"distance_cm\": 4.9}, {\"object\": \"faucet\", \"relation\": \"side_above\", \"distance_cm\": 38.7}]}", "target": "hair dryer", "split": "train"}
{"id": "scene0250_00_4", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0250_00_5", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bathroom cabinet\nB. bottle\nC. bathtub\nD. toilet", "choices": ["A. bathroom cabinet", "B. bottle", "C. bathtub", "D. toilet"], "answer": "D. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0250_00_6", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 27 cm away. It is also to the side of and above the rug, about 63 cm away. It is further to the side of and below the bottle, about 75 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0250_00_7", "scene_id": "scene0250_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"rug\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 27.3}, {\"object\": \"rug\", \"relation\": \"side_above\", \"distance_cm\": 63.0}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0257_00_0", "scene_id": "scene0257_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0257_00_1", "scene_id": "scene0257_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the fire extinguisher in this scene?\nA. sink\nB. monitor\nC. desk\nD. chair", "choices": ["A. sink", "B. monitor", "C. desk", "D. chair"], "answer": "A. sink", "target": "fire extinguisher", "split": "train"}
{"id": "scene0257_00_2", "scene_id": "scene0257_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the fire extinguisher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The fire extinguisher is to the side of and below the sink, about 16 cm away. It is also near the chair, about 36 cm away.", "target": "fire extinguisher", "split": "train"}
{"id": "scene0257_00_3", "scene_id": "scene0257_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the fire extinguisher in structured JSON format.", "choices": null, "answer": "{\"target\": \"fire extinguisher\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 15.9}, {\"object\": \"chair\", \"relation\": \"near\", \"distance_cm\": 36.2}]}", "target": "fire extinguisher", "split": "train"}
{"id": "scene0260_00_0", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "Yes", "target": "hair dryer", "split": "train"}
{"id": "scene0260_00_1", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hair dryer in this scene?\nA. towel\nB. toilet paper\nC. toilet\nD. mirror", "choices": ["A. towel", "B. toilet paper", "C. toilet", "D. mirror"], "answer": "A. towel", "target": "hair dryer", "split": "train"}
{"id": "scene0260_00_2", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hair dryer in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hair dryer is to the side of and below the towel, about 20 cm away. It is also to the side of and below the mirror, about 31 cm away. It is further to the side of and above the toilet paper, about 57 cm away.", "target": "hair dryer", "split": "train"}
{"id": "scene0260_00_3", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hair dryer in structured JSON format.", "choices": null, "answer": "{\"target\": \"hair dryer\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 19.8}, {\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 31.4}]}", "target": "hair dryer", "split": "train"}
{"id": "scene0260_00_4", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "No", "target": "telephone", "split": "train"}
{"id": "scene0260_00_5", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bathroom vanity\nB. toilet\nC. towel\nD. hair dryer", "choices": ["A. bathroom vanity", "B. toilet", "C. towel", "D. hair dryer"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0260_00_6", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the toilet, about 17 cm away. It is also to the side of and below the bathroom vanity, about 24 cm away. It is further to the side of and below the towel, about 46 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0260_00_7", "scene_id": "scene0260_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"trash can\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 17.2}, {\"object\": \"bathroom vanity\", \"relation\": \"side_below\", \"distance_cm\": 24.0}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0261_00_0", "scene_id": "scene0261_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0261_00_1", "scene_id": "scene0261_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. chair\nB. toilet\nC. monitor\nD. ironing board", "choices": ["A. chair", "B. toilet", "C. monitor", "D. ironing board"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0261_00_2", "scene_id": "scene0261_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 12 cm away. It is also to the side of and below the ironing board, about 32 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0261_00_3", "scene_id": "scene0261_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 11.7}, {\"object\": \"ironing board\", \"relation\": \"side_below\", \"distance_cm\": 31.9}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0263_00_0", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0263_00_1", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. telephone\nC. computer tower\nD. monitor", "choices": ["A. chair", "B. telephone", "C. computer tower", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0263_00_2", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and below the monitor, about 6 cm away. It is further near the telephone, about 17 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0263_00_3", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.4}, {\"object\": \"telephone\", \"relation\": \"near\", \"distance_cm\": 16.8}]}", "target": "keyboard", "split": "train"}
{"id": "scene0263_00_4", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "train"}
{"id": "scene0263_00_5", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. chair\nB. keyboard\nC. computer tower\nD. monitor", "choices": ["A. chair", "B. keyboard", "C. computer tower", "D. monitor"], "answer": "D. monitor", "target": "telephone", "split": "train"}
{"id": "scene0263_00_6", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also to the side of and below the monitor, about 13 cm away. It is further near the keyboard, about 17 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0263_00_7", "scene_id": "scene0263_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.3}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 16.8}]}", "target": "telephone", "split": "train"}
{"id": "scene0265_00_0", "scene_id": "scene0265_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0265_00_1", "scene_id": "scene0265_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. toilet\nB. towel\nC. trash can\nD. bathroom vanity", "choices": ["A. toilet", "B. towel", "C. trash can", "D. bathroom vanity"], "answer": "C. trash can", "target": "toilet paper", "split": "train"}
{"id": "scene0265_00_2", "scene_id": "scene0265_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is above the trash can, about 11 cm away. It is also near the toilet, about 18 cm away. It is further to the side of and below the bathroom vanity, about 31 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0265_00_3", "scene_id": "scene0265_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"trash can\", \"references\": [{\"object\": \"trash can\", \"relation\": \"above\", \"distance_cm\": 11.0}, {\"object\": \"toilet\", \"relation\": \"near\", \"distance_cm\": 17.7}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0266_00_0", "scene_id": "scene0266_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0266_00_1", "scene_id": "scene0266_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0268_00_0", "scene_id": "scene0268_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0268_00_1", "scene_id": "scene0268_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. cup\nB. bathroom counter\nC. sink\nD. toilet", "choices": ["A. cup", "B. bathroom counter", "C. sink", "D. toilet"], "answer": "A. cup", "target": "tissue box", "split": "train"}
{"id": "scene0268_00_2", "scene_id": "scene0268_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is next to the cup, about 8 cm away. It is also to the side of and above the sink, about 20 cm away. It is further to the side of and above the toilet, about 67 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0268_00_3", "scene_id": "scene0268_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 7.6}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 20.4}]}", "target": "tissue box", "split": "train"}
{"id": "scene0270_00_0", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0270_00_1", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the clock in this scene?\nA. cabinet\nB. chair\nC. backpack\nD. pillow", "choices": ["A. cabinet", "B. chair", "C. backpack", "D. pillow"], "answer": "D. pillow", "target": "clock", "split": "train"}
{"id": "scene0270_00_2", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The clock is on the nightstand. It is also to the side of and below the pillow, about 33 cm away.", "target": "clock", "split": "train"}
{"id": "scene0270_00_3", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"clock\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 32.5}]}", "target": "clock", "split": "train"}
{"id": "scene0270_00_4", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "Yes", "target": "paper bag", "split": "train"}
{"id": "scene0270_00_5", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. backpack\nB. chair\nC. cabinet\nD. luggage rack", "choices": ["A. backpack", "B. chair", "C. cabinet", "D. luggage rack"], "answer": "D. luggage rack", "target": "paper bag", "split": "train"}
{"id": "scene0270_00_6", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is to the side of and below the luggage rack, about 16 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0270_00_7", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"luggage rack\", \"relation\": \"side_below\", \"distance_cm\": 15.8}]}", "target": "paper bag", "split": "train"}
{"id": "scene0270_00_8", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper towel roll appear?", "choices": null, "answer": "No", "target": "paper towel roll", "split": "train"}
{"id": "scene0270_00_9", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. cabinet\nB. chair\nC. mirror\nD. backpack", "choices": ["A. cabinet", "B. chair", "C. mirror", "D. backpack"], "answer": "C. mirror", "target": "telephone", "split": "train"}
{"id": "scene0270_00_10", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also to the side of and below the mirror, about 34 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0270_00_11", "scene_id": "scene0270_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 33.8}]}", "target": "telephone", "split": "train"}
{"id": "scene0271_00_0", "scene_id": "scene0271_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "scene0271_00_1", "scene_id": "scene0271_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the purse in this scene?\nA. plant\nB. dresser\nC. backpack\nD. picture", "choices": ["A. plant", "B. dresser", "C. backpack", "D. picture"], "answer": "A. plant", "target": "purse", "split": "train"}
{"id": "scene0271_00_2", "scene_id": "scene0271_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the purse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The purse is on the dresser. It is also to the side of and below the plant, about 13 cm away. It is further to the side of and above the backpack, about 46 cm away.", "target": "purse", "split": "train"}
{"id": "scene0271_00_3", "scene_id": "scene0271_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the purse in structured JSON format.", "choices": null, "answer": "{\"target\": \"purse\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 13.5}, {\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 46.1}]}", "target": "purse", "split": "train"}
{"id": "scene0276_00_0", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "train"}
{"id": "scene0276_00_1", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. towel\nB. mirror\nC. bathroom cabinet\nD. sink", "choices": ["A. towel", "B. mirror", "C. bathroom cabinet", "D. sink"], "answer": "A. towel", "target": "telephone", "split": "train"}
{"id": "scene0276_00_2", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the towel, about 15 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0276_00_3", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"towel\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 14.7}]}", "target": "telephone", "split": "train"}
{"id": "scene0276_00_4", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "No", "target": "bag", "split": "train"}
{"id": "scene0276_00_5", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. mirror\nB. toilet\nC. bathroom cabinet\nD. sink", "choices": ["A. mirror", "B. toilet", "C. bathroom cabinet", "D. sink"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0276_00_6", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 19 cm away. It is also to the side of and below the sink, about 28 cm away. It is further to the side of and below the mirror, about 56 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0276_00_7", "scene_id": "scene0276_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 18.5}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 28.1}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0279_00_0", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "Yes", "target": "alarm clock", "split": "train"}
{"id": "scene0279_00_1", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the alarm clock in this scene?\nA. backpack\nB. lamp\nC. pillow\nD. bag", "choices": ["A. backpack", "B. lamp", "C. pillow", "D. bag"], "answer": "C. pillow", "target": "alarm clock", "split": "train"}
{"id": "scene0279_00_2", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the alarm clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The alarm clock is on the nightstand. It is also to the side of and below the pillow, about 44 cm away. It is further to the side of and below the lamp, about 45 cm away.", "target": "alarm clock", "split": "train"}
{"id": "scene0279_00_3", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the alarm clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"alarm clock\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 44.3}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 45.3}]}", "target": "alarm clock", "split": "train"}
{"id": "scene0279_00_4", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0279_00_5", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. lamp\nB. bag\nC. alarm clock\nD. nightstand", "choices": ["A. lamp", "B. bag", "C. alarm clock", "D. nightstand"], "answer": "B. bag", "target": "backpack", "split": "train"}
{"id": "scene0279_00_6", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the bag, about 8 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0279_00_7", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 8.3}]}", "target": "backpack", "split": "train"}
{"id": "scene0279_00_8", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0279_00_9", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. suitcase\nB. lamp\nC. alarm clock\nD. backpack", "choices": ["A. suitcase", "B. lamp", "C. alarm clock", "D. backpack"], "answer": "D. backpack", "target": "bag", "split": "train"}
{"id": "scene0279_00_10", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the backpack, about 8 cm away. It is also to the side of and below the suitcase, about 27 cm away.", "target": "bag", "split": "train"}
{"id": "scene0279_00_11", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"backpack\", \"references\": [{\"object\": \"backpack\", \"relation\": \"side_below\", \"distance_cm\": 8.3}, {\"object\": \"suitcase\", \"relation\": \"side_below\", \"distance_cm\": 27.1}]}", "target": "bag", "split": "train"}
{"id": "scene0279_00_12", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0279_00_13", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. alarm clock\nB. lamp\nC. backpack\nD. suitcase", "choices": ["A. alarm clock", "B. lamp", "C. backpack", "D. suitcase"], "answer": "B. lamp", "target": "telephone", "split": "train"}
{"id": "scene0279_00_14", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the nightstand. It is also to the side of and below the lamp, about 43 cm away. It is further to the side of and above the suitcase, about 78 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0279_00_15", "scene_id": "scene0279_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 42.8}, {\"object\": \"suitcase\", \"relation\": \"side_above\", \"distance_cm\": 78.2}]}", "target": "telephone", "split": "train"}
{"id": "scene0280_00_0", "scene_id": "scene0280_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0280_00_1", "scene_id": "scene0280_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. monitor\nB. desk\nC. box\nD. chair", "choices": ["A. monitor", "B. desk", "C. box", "D. chair"], "answer": "C. box", "target": "ball", "split": "train"}
{"id": "scene0280_00_2", "scene_id": "scene0280_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is next to the box, about 5 cm away.", "target": "ball", "split": "train"}
{"id": "scene0280_00_3", "scene_id": "scene0280_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"box\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 4.8}]}", "target": "ball", "split": "train"}
{"id": "scene0282_00_0", "scene_id": "scene0282_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "train"}
{"id": "scene0282_00_1", "scene_id": "scene0282_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. shelf\nB. monitor\nC. bed\nD. fan", "choices": ["A. shelf", "B. monitor", "C. bed", "D. fan"], "answer": "C. bed", "target": "cup", "split": "train"}
{"id": "scene0282_00_2", "scene_id": "scene0282_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the bed, about 14 cm away. It is also to the side of and above the fan, about 54 cm away.", "target": "cup", "split": "train"}
{"id": "scene0282_00_3", "scene_id": "scene0282_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bed\", \"relation\": \"next_to\", \"distance_cm\": 13.5}, {\"object\": \"fan\", \"relation\": \"side_above\", \"distance_cm\": 54.3}]}", "target": "cup", "split": "train"}
{"id": "scene0286_00_0", "scene_id": "scene0286_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0286_00_1", "scene_id": "scene0286_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. chair\nB. lamp\nC. nightstand\nD. monitor", "choices": ["A. chair", "B. lamp", "C. nightstand", "D. monitor"], "answer": "B. lamp", "target": "book", "split": "train"}
{"id": "scene0286_00_2", "scene_id": "scene0286_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the nightstand. It is also to the side of and below the lamp, about 11 cm away.", "target": "book", "split": "train"}
{"id": "scene0286_00_3", "scene_id": "scene0286_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 10.8}]}", "target": "book", "split": "train"}
{"id": "scene0298_00_0", "scene_id": "scene0298_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "No", "target": "book", "split": "train"}
{"id": "scene0299_00_0", "scene_id": "scene0299_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plunger appear?", "choices": null, "answer": "No", "target": "plunger", "split": "train"}
{"id": "scene0300_00_0", "scene_id": "scene0300_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0300_00_1", "scene_id": "scene0300_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. table\nB. chair\nC. trash can\nD. divider", "choices": ["A. table", "B. chair", "C. trash can", "D. divider"], "answer": "D. divider", "target": "backpack", "split": "train"}
{"id": "scene0300_00_2", "scene_id": "scene0300_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and below the divider, about 27 cm away. It is further to the side of and above the trash can, about 53 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0300_00_3", "scene_id": "scene0300_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"divider\", \"relation\": \"side_below\", \"distance_cm\": 27.3}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 52.7}]}", "target": "backpack", "split": "train"}
{"id": "scene0301_00_0", "scene_id": "scene0301_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0301_00_1", "scene_id": "scene0301_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. plant\nB. chair\nC. desk\nD. monitor", "choices": ["A. plant", "B. chair", "C. desk", "D. monitor"], "answer": "A. plant", "target": "ball", "split": "train"}
{"id": "scene0301_00_2", "scene_id": "scene0301_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is to the side of and below the plant, about 45 cm away.", "target": "ball", "split": "train"}
{"id": "scene0301_00_3", "scene_id": "scene0301_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"plant\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 45.3}]}", "target": "ball", "split": "train"}
{"id": "scene0302_00_0", "scene_id": "scene0302_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plunger appear?", "choices": null, "answer": "No", "target": "plunger", "split": "train"}
{"id": "scene0302_00_1", "scene_id": "scene0302_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. keyboard\nC. monitor\nD. desk", "choices": ["A. chair", "B. keyboard", "C. monitor", "D. desk"], "answer": "A. chair", "target": "backpack", "split": "train"}
{"id": "scene0302_00_2", "scene_id": "scene0302_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is under the chair, about 3 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0302_00_3", "scene_id": "scene0302_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"under\", \"distance_cm\": 3.4}]}", "target": "backpack", "split": "train"}
{"id": "scene0303_00_0", "scene_id": "scene0303_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "No", "target": "bottle", "split": "train"}
{"id": "scene0303_00_1", "scene_id": "scene0303_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper cutter in this scene?\nA. monitor\nB. picture\nC. counter\nD. cabinet", "choices": ["A. monitor", "B. picture", "C. counter", "D. cabinet"], "answer": "D. cabinet", "target": "paper cutter", "split": "train"}
{"id": "scene0303_00_2", "scene_id": "scene0303_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper cutter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper cutter is to the side of and above the cabinet, about 6 cm away. It is also to the side of and below the picture, about 40 cm away.", "target": "paper cutter", "split": "train"}
{"id": "scene0303_00_3", "scene_id": "scene0303_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper cutter in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper cutter\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 6.4}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 39.6}]}", "target": "paper cutter", "split": "train"}
{"id": "scene0306_00_0", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0306_00_1", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. computer tower\nB. laptop\nC. paper bag\nD. chair", "choices": ["A. computer tower", "B. laptop", "C. paper bag", "D. chair"], "answer": "A. computer tower", "target": "keyboard", "split": "train"}
{"id": "scene0306_00_2", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and above the computer tower, about 20 cm away. It is further to the side of and below the laptop, about 26 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0306_00_3", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 20.2}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 26.3}]}", "target": "keyboard", "split": "train"}
{"id": "scene0306_00_4", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0306_00_5", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. desk\nB. trash can\nC. paper bag\nD. keyboard", "choices": ["A. desk", "B. trash can", "C. paper bag", "D. keyboard"], "answer": "A. desk", "target": "laptop", "split": "train"}
{"id": "scene0306_00_6", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is to the side of and above the desk, about 3 cm away. It is also to the side of and above the keyboard, about 26 cm away. It is further to the side of and above the paper bag, about 33 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0306_00_7", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 2.7}, {\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 26.3}]}", "target": "laptop", "split": "train"}
{"id": "scene0306_00_8", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0306_00_9", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. trash can\nB. laptop\nC. keyboard\nD. chair", "choices": ["A. trash can", "B. laptop", "C. keyboard", "D. chair"], "answer": "A. trash can", "target": "paper bag", "split": "train"}
{"id": "scene0306_00_10", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is to the side of and above the trash can, about 6 cm away. It is also to the side of and below the laptop, about 33 cm away. It is further to the side of and below the keyboard, about 37 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0306_00_11", "scene_id": "scene0306_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 6.3}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 32.5}]}", "target": "paper bag", "split": "train"}
{"id": "scene0309_00_0", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0309_00_1", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. display rack\nB. person\nC. vacuum cleaner\nD. chair", "choices": ["A. display rack", "B. person", "C. vacuum cleaner", "D. chair"], "answer": "C. vacuum cleaner", "target": "bag", "split": "train"}
{"id": "scene0309_00_2", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the vacuum cleaner, about 36 cm away. It is also to the side of and above the person, about 57 cm away.", "target": "bag", "split": "train"}
{"id": "scene0309_00_3", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"vacuum cleaner\", \"references\": [{\"object\": \"vacuum cleaner\", \"relation\": \"side_above\", \"distance_cm\": 36.2}, {\"object\": \"person\", \"relation\": \"side_above\", \"distance_cm\": 56.5}]}", "target": "bag", "split": "train"}
{"id": "scene0309_00_4", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "train"}
{"id": "scene0309_00_5", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. person\nB. chair\nC. display rack\nD. monitor", "choices": ["A. person", "B. chair", "C. display rack", "D. monitor"], "answer": "D. monitor", "target": "telephone", "split": "train"}
{"id": "scene0309_00_6", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is next to the monitor, about 10 cm away. It is also to the side of and above the chair, about 40 cm away. It is further to the side of and above the person, about 60 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0309_00_7", "scene_id": "scene0309_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"display rack\", \"references\": [{\"object\": \"monitor\", \"relation\": \"next_to\", \"distance_cm\": 10.1}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 40.1}]}", "target": "telephone", "split": "train"}
{"id": "scene0311_00_0", "scene_id": "scene0311_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the candle appear?", "choices": null, "answer": "No", "target": "candle", "split": "train"}
{"id": "scene0311_00_1", "scene_id": "scene0311_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "Yes", "target": "paper cutter", "split": "train"}
{"id": "scene0312_00_0", "scene_id": "scene0312_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0312_00_1", "scene_id": "scene0312_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. jacket\nB. monitor\nC. lamp\nD. table", "choices": ["A. jacket", "B. monitor", "C. lamp", "D. table"], "answer": "D. table", "target": "backpack", "split": "train"}
{"id": "scene0312_00_2", "scene_id": "scene0312_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the table, about 13 cm away. It is also to the side of and below the lamp, about 30 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0312_00_3", "scene_id": "scene0312_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 12.9}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 29.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0320_00_0", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0320_00_1", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. book\nB. chair\nC. bookshelves\nD. books", "choices": ["A. book", "B. chair", "C. bookshelves", "D. books"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0320_00_2", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 43 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0320_00_3", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 43.2}]}", "target": "backpack", "split": "train"}
{"id": "scene0320_00_4", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0320_00_5", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. chair\nB. books\nC. backpack\nD. bookshelves", "choices": ["A. chair", "B. books", "C. backpack", "D. bookshelves"], "answer": "B. books", "target": "book", "split": "train"}
{"id": "scene0320_00_6", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is to the side of and above the books, about 39 cm away.", "target": "book", "split": "train"}
{"id": "scene0320_00_7", "scene_id": "scene0320_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"books\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 39.1}]}", "target": "book", "split": "train"}
{"id": "scene0322_00_0", "scene_id": "scene0322_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dustpan appear?", "choices": null, "answer": "No", "target": "dustpan", "split": "train"}
{"id": "scene0322_00_1", "scene_id": "scene0322_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. monitor\nB. table\nC. desk\nD. chair", "choices": ["A. monitor", "B. table", "C. desk", "D. chair"], "answer": "D. chair", "target": "cup", "split": "train"}
{"id": "scene0322_00_2", "scene_id": "scene0322_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and above the chair, about 29 cm away.", "target": "cup", "split": "train"}
{"id": "scene0322_00_3", "scene_id": "scene0322_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 28.5}]}", "target": "cup", "split": "train"}
{"id": "scene0325_00_0", "scene_id": "scene0325_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0325_00_1", "scene_id": "scene0325_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. computer tower\nB. table\nC. file cabinet\nD. printer", "choices": ["A. computer tower", "B. table", "C. file cabinet", "D. printer"], "answer": "B. table", "target": "power strip", "split": "train"}
{"id": "scene0325_00_2", "scene_id": "scene0325_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the table, about 32 cm away. It is also to the side of and below the file cabinet, about 52 cm away. It is further to the side of and below the printer, about 70 cm away.", "target": "power strip", "split": "train"}
{"id": "scene0325_00_3", "scene_id": "scene0325_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 32.1}, {\"object\": \"file cabinet\", \"relation\": \"side_below\", \"distance_cm\": 51.5}]}", "target": "power strip", "split": "train"}
{"id": "scene0332_00_0", "scene_id": "scene0332_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0335_00_0", "scene_id": "scene0335_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0340_00_0", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "train"}
{"id": "scene0340_00_1", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bed\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0340_00_2", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0340_00_3", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. chair\nB. trash can\nC. picture\nD. lamp", "choices": ["A. chair", "B. trash can", "C. picture", "D. lamp"], "answer": "C. picture", "target": "laptop", "split": "train"}
{"id": "scene0340_00_4", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also to the side of and below the picture, about 28 cm away. It is further to the side of and above the chair, about 35 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0340_00_5", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 27.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 35.3}]}", "target": "laptop", "split": "train"}
{"id": "scene0340_00_6", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0340_00_7", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. lamp\nB. pillow\nC. bed\nD. backpack", "choices": ["A. lamp", "B. pillow", "C. bed", "D. backpack"], "answer": "B. pillow", "target": "telephone", "split": "train"}
{"id": "scene0340_00_8", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the nightstand. It is also near the pillow, about 30 cm away. It is further to the side of and below the lamp, about 43 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0340_00_9", "scene_id": "scene0340_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"pillow\", \"relation\": \"near\", \"distance_cm\": 29.7}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 43.2}]}", "target": "telephone", "split": "train"}
{"id": "scene0341_00_0", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0341_00_1", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothing\nB. computer tower\nC. file cabinet\nD. office chair", "choices": ["A. clothing", "B. computer tower", "C. file cabinet", "D. office chair"], "answer": "A. clothing", "target": "backpack", "split": "train"}
{"id": "scene0341_00_2", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the clothing, about 13 cm away. It is also to the side of and below the office chair, about 14 cm away. It is further to the side of and below the computer tower, about 16 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0341_00_3", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"couch\", \"references\": [{\"object\": \"clothing\", \"relation\": \"side_below\", \"distance_cm\": 12.6}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 14.1}]}", "target": "backpack", "split": "train"}
{"id": "scene0341_00_4", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "Yes", "target": "ball", "split": "train"}
{"id": "scene0341_00_5", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. backpack\nB. trash can\nC. plant\nD. bookshelf", "choices": ["A. backpack", "B. trash can", "C. plant", "D. bookshelf"], "answer": "B. trash can", "target": "ball", "split": "train"}
{"id": "scene0341_00_6", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is to the side of and below the trash can, about 48 cm away. It is also to the side of and below the plant, about 54 cm away.", "target": "ball", "split": "train"}
{"id": "scene0341_00_7", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 48.0}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 54.5}]}", "target": "ball", "split": "train"}
{"id": "scene0341_00_8", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0341_00_9", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. file cabinet\nB. keyboard\nC. cup\nD. monitor", "choices": ["A. file cabinet", "B. keyboard", "C. cup", "D. monitor"], "answer": "A. file cabinet", "target": "headphones", "split": "train"}
{"id": "scene0341_00_10", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the file cabinet, about 6 cm away. It is also to the side of and below the monitor, about 31 cm away. It is further near the keyboard, about 35 cm away.", "target": "headphones", "split": "train"}
{"id": "scene0341_00_11", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 5.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 30.9}]}", "target": "headphones", "split": "train"}
{"id": "scene0341_00_12", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "Yes", "target": "paper bag", "split": "train"}
{"id": "scene0341_00_13", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. box\nB. backpack\nC. bookshelf\nD. desk", "choices": ["A. box", "B. backpack", "C. bookshelf", "D. desk"], "answer": "D. desk", "target": "paper bag", "split": "train"}
{"id": "scene0341_00_14", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is to the side of and below the desk, about 10 cm away. It is also to the side of and below the bookshelf, about 12 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0341_00_15", "scene_id": "scene0341_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 9.9}, {\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 11.8}]}", "target": "paper bag", "split": "train"}
{"id": "scene0346_00_0", "scene_id": "scene0346_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "train"}
{"id": "scene0346_00_1", "scene_id": "scene0346_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. toilet\nB. mirror\nC. bathroom cabinet\nD. sink", "choices": ["A. toilet", "B. mirror", "C. bathroom cabinet", "D. sink"], "answer": "D. sink", "target": "toilet paper", "split": "train"}
{"id": "scene0346_00_2", "scene_id": "scene0346_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the sink, about 32 cm away. It is also to the side of and above the toilet, about 41 cm away. It is further to the side of and below the mirror, about 61 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0346_00_3", "scene_id": "scene0346_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 31.6}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 41.4}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0348_00_0", "scene_id": "scene0348_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "train"}
{"id": "scene0348_00_1", "scene_id": "scene0348_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. cabinet\nB. cutting board\nC. monitor\nD. kitchen counter", "choices": ["A. cabinet", "B. cutting board", "C. monitor", "D. kitchen counter"], "answer": "B. cutting board", "target": "cup", "split": "train"}
{"id": "scene0348_00_2", "scene_id": "scene0348_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the cutting board, about 15 cm away. It is also to the side of and below the cabinet, about 44 cm away.", "target": "cup", "split": "train"}
{"id": "scene0348_00_3", "scene_id": "scene0348_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"cutting board\", \"relation\": \"next_to\", \"distance_cm\": 14.9}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 43.7}]}", "target": "cup", "split": "train"}
{"id": "scene0351_00_0", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0351_00_1", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0351_00_2", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. monitor\nB. backpack\nC. chair\nD. plant", "choices": ["A. monitor", "B. backpack", "C. chair", "D. plant"], "answer": "D. plant", "target": "book", "split": "train"}
{"id": "scene0351_00_3", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the desk. It is also to the side of and below the plant, about 48 cm away. It is further to the side of and above the chair, about 49 cm away.", "target": "book", "split": "train"}
{"id": "scene0351_00_4", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 47.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 48.9}]}", "target": "book", "split": "train"}
{"id": "scene0351_00_5", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0351_00_6", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. radiator\nB. backpack\nC. monitor\nD. keyboard", "choices": ["A. radiator", "B. backpack", "C. monitor", "D. keyboard"], "answer": "C. monitor", "target": "telephone", "split": "train"}
{"id": "scene0351_00_7", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and below the monitor, about 9 cm away. It is also near the keyboard, about 34 cm away. It is further to the side of and above the radiator, about 53 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0351_00_8", "scene_id": "scene0351_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"monitor\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.0}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 34.2}]}", "target": "telephone", "split": "train"}
{"id": "scene0356_00_0", "scene_id": "scene0356_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "scene0356_00_1", "scene_id": "scene0356_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. desk\nC. chair\nD. dresser", "choices": ["A. monitor", "B. desk", "C. chair", "D. dresser"], "answer": "D. dresser", "target": "telephone", "split": "train"}
{"id": "scene0356_00_2", "scene_id": "scene0356_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the dresser, about 3 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0356_00_3", "scene_id": "scene0356_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"dresser\", \"relation\": \"side_above\", \"distance_cm\": 3.1}]}", "target": "telephone", "split": "train"}
{"id": "scene0357_00_0", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "train"}
{"id": "scene0357_00_1", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. pipe\nC. cabinet\nD. shoes", "choices": ["A. bag", "B. pipe", "C. cabinet", "D. shoes"], "answer": "D. shoes", "target": "backpack", "split": "train"}
{"id": "scene0357_00_2", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the shoes, about 5 cm away. It is also to the side of and below the cabinet, about 53 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0357_00_3", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 4.6}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 53.0}]}", "target": "backpack", "split": "train"}
{"id": "scene0357_00_4", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0357_00_5", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. backpack\nB. towel\nC. cabinet\nD. shoes", "choices": ["A. backpack", "B. towel", "C. cabinet", "D. shoes"], "answer": "B. towel", "target": "bag", "split": "train"}
{"id": "scene0357_00_6", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is on the cabinet. It is also to the side of and below the towel, about 9 cm away. It is further to the side of and above the shoes, about 39 cm away.", "target": "bag", "split": "train"}
{"id": "scene0357_00_7", "scene_id": "scene0357_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 8.6}, {\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 38.9}]}", "target": "bag", "split": "train"}
{"id": "scene0359_00_0", "scene_id": "scene0359_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the fire extinguisher appear?", "choices": null, "answer": "No", "target": "fire extinguisher", "split": "train"}
{"id": "scene0359_00_1", "scene_id": "scene0359_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper towel roll in this scene?\nA. desk\nB. shelf\nC. spray bottle\nD. monitor", "choices": ["A. desk", "B. shelf", "C. spray bottle", "D. monitor"], "answer": "C. spray bottle", "target": "paper towel roll", "split": "train"}
{"id": "scene0359_00_2", "scene_id": "scene0359_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper towel roll in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper towel roll is on the desk. It is also near the spray bottle, about 16 cm away.", "target": "paper towel roll", "split": "train"}
{"id": "scene0359_00_3", "scene_id": "scene0359_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"spray bottle\", \"relation\": \"near\", \"distance_cm\": 16.2}]}", "target": "paper towel roll", "split": "train"}
{"id": "scene0365_00_0", "scene_id": "scene0365_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0365_00_1", "scene_id": "scene0365_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. bar\nB. sink\nC. monitor\nD. mirror", "choices": ["A. bar", "B. sink", "C. monitor", "D. mirror"], "answer": "B. sink", "target": "soap dispenser", "split": "train"}
{"id": "scene0365_00_2", "scene_id": "scene0365_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the sink, about 5 cm away. It is also to the side of and below the bar, about 35 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0365_00_3", "scene_id": "scene0365_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"sink\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 4.8}, {\"object\": \"bar\", \"relation\": \"side_below\", \"distance_cm\": 34.5}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0366_00_0", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dumbbell appear?", "choices": null, "answer": "No", "target": "dumbbell", "split": "train"}
{"id": "scene0366_00_1", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plunger in this scene?\nA. scale\nB. sink\nC. toilet paper\nD. toilet", "choices": ["A. scale", "B. sink", "C. toilet paper", "D. toilet"], "answer": "D. toilet", "target": "plunger", "split": "train"}
{"id": "scene0366_00_2", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plunger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plunger is to the side of and below the toilet, about 12 cm away.", "target": "plunger", "split": "train"}
{"id": "scene0366_00_3", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plunger in structured JSON format.", "choices": null, "answer": "{\"target\": \"plunger\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 12.1}]}", "target": "plunger", "split": "train"}
{"id": "scene0366_00_4", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "Yes", "target": "scale", "split": "train"}
{"id": "scene0366_00_5", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scale in this scene?\nA. sink\nB. toilet\nC. trash can\nD. plunger", "choices": ["A. sink", "B. toilet", "C. trash can", "D. plunger"], "answer": "C. trash can", "target": "scale", "split": "train"}
{"id": "scene0366_00_6", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is to the side of and below the trash can, about 11 cm away. It is also to the side of and below the sink, about 65 cm away.", "target": "scale", "split": "train"}
{"id": "scene0366_00_7", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"trash can\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 10.7}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 65.3}]}", "target": "scale", "split": "train"}
{"id": "scene0366_00_8", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "scene0366_00_9", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. trash can\nB. plunger\nC. toilet\nD. scale", "choices": ["A. trash can", "B. plunger", "C. toilet", "D. scale"], "answer": "C. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0366_00_10", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 36 cm away. It is also to the side of and above the trash can, about 73 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0366_00_11", "scene_id": "scene0366_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 35.7}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 73.3}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0372_00_0", "scene_id": "scene0372_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0372_00_1", "scene_id": "scene0372_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. keyboard\nB. desk\nC. monitor\nD. chair", "choices": ["A. keyboard", "B. desk", "C. monitor", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "train"}
{"id": "scene0374_00_0", "scene_id": "scene0374_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "train"}
{"id": "scene0374_00_1", "scene_id": "scene0374_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"couch\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0375_00_0", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0375_00_1", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. mirror\nB. trash can\nC. sink\nD. toilet paper", "choices": ["A. mirror", "B. trash can", "C. sink", "D. toilet paper"], "answer": "A. mirror", "target": "bag", "split": "train"}
{"id": "scene0375_00_2", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the mirror, about 7 cm away. It is also to the side of and above the sink, about 13 cm away. It is further to the side of and above the trash can, about 57 cm away.", "target": "bag", "split": "train"}
{"id": "scene0375_00_3", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 7.4}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 13.5}]}", "target": "bag", "split": "train"}
{"id": "scene0375_00_4", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "Yes", "target": "soap dish", "split": "train"}
{"id": "scene0375_00_5", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. bar\nB. bathroom vanity\nC. bag\nD. mirror", "choices": ["A. bar", "B. bathroom vanity", "C. bag", "D. mirror"], "answer": "A. bar", "target": "soap dish", "split": "train"}
{"id": "scene0375_00_6", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is near the bar, about 31 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0375_00_7", "scene_id": "scene0375_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"shower\", \"references\": [{\"object\": \"bar\", \"relation\": \"near\", \"distance_cm\": 31.2}]}", "target": "soap dish", "split": "train"}
{"id": "scene0378_00_0", "scene_id": "scene0378_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0379_00_0", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0379_00_1", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. file cabinet\nB. computer tower\nC. monitor\nD. chair", "choices": ["A. file cabinet", "B. computer tower", "C. monitor", "D. chair"], "answer": "D. chair", "target": "bottle", "split": "train"}
{"id": "scene0379_00_2", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the chair, about 13 cm away. It is further to the side of and above the file cabinet, about 15 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0379_00_3", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 12.5}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 15.0}]}", "target": "bottle", "split": "train"}
{"id": "scene0379_00_4", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "scene0379_00_5", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. bottle\nB. computer tower\nC. monitor\nD. chair", "choices": ["A. bottle", "B. computer tower", "C. monitor", "D. chair"], "answer": "C. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0379_00_6", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is under the monitor, about 4 cm away. It is also to the side of and below the chair, about 23 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0379_00_7", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.5}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 23.0}]}", "target": "keyboard", "split": "train"}
{"id": "scene0379_00_8", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0379_00_9", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. chair\nB. table\nC. keyboard\nD. monitor", "choices": ["A. chair", "B. table", "C. keyboard", "D. monitor"], "answer": "D. monitor", "target": "telephone", "split": "train"}
{"id": "scene0379_00_10", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and below the monitor, about 12 cm away. It is also to the side of and above the chair, about 17 cm away. It is further near the keyboard, about 18 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0379_00_11", "scene_id": "scene0379_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 17.2}]}", "target": "telephone", "split": "train"}
{"id": "scene0380_00_0", "scene_id": "scene0380_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0380_00_1", "scene_id": "scene0380_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. desk\nB. monitor\nC. keyboard\nD. chair", "choices": ["A. desk", "B. monitor", "C. keyboard", "D. chair"], "answer": "B. monitor", "target": "telephone", "split": "train"}
{"id": "scene0380_00_2", "scene_id": "scene0380_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is on the desk. It is also to the side of and below the monitor, about 27 cm away. It is further near the keyboard, about 34 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0380_00_3", "scene_id": "scene0380_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 27.0}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 34.2}]}", "target": "telephone", "split": "train"}
{"id": "scene0383_00_0", "scene_id": "scene0383_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the shoe appear?", "choices": null, "answer": "No", "target": "shoe", "split": "train"}
{"id": "scene0383_00_1", "scene_id": "scene0383_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. chair\nB. radiator\nC. monitor\nD. desk", "choices": ["A. chair", "B. radiator", "C. monitor", "D. desk"], "answer": "B. radiator", "target": "telephone", "split": "train"}
{"id": "scene0383_00_2", "scene_id": "scene0383_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the radiator, about 28 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0383_00_3", "scene_id": "scene0383_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"radiator\", \"references\": [{\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 27.7}]}", "target": "telephone", "split": "train"}
{"id": "scene0384_00_0", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0384_00_1", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. shoe rack\nB. pipe\nC. shoes\nD. bicycle", "choices": ["A. shoe rack", "B. pipe", "C. shoes", "D. bicycle"], "answer": "D. bicycle", "target": "bag", "split": "train"}
{"id": "scene0384_00_2", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is under the bicycle, about 3 cm away. It is also to the side of and below the pipe, about 19 cm away. It is further near the shoe rack, about 33 cm away.", "target": "bag", "split": "train"}
{"id": "scene0384_00_3", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"shoe rack\", \"references\": [{\"object\": \"bicycle\", \"relation\": \"under\", \"distance_cm\": 3.5}, {\"object\": \"pipe\", \"relation\": \"side_below\", \"distance_cm\": 19.3}]}", "target": "bag", "split": "train"}
{"id": "scene0384_00_4", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the ball appear?", "choices": null, "answer": "Yes", "target": "ball", "split": "train"}
{"id": "scene0384_00_5", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. box\nB. bag\nC. case\nD. shoes", "choices": ["A. box", "B. bag", "C. case", "D. shoes"], "answer": "C. case", "target": "ball", "split": "train"}
{"id": "scene0384_00_6", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is next to the case, about 4 cm away. It is also next to the bag, about 8 cm away. It is further to the side of and below the box, about 28 cm away.", "target": "ball", "split": "train"}
{"id": "scene0384_00_7", "scene_id": "scene0384_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"case\", \"references\": [{\"object\": \"case\", \"relation\": \"next_to\", \"distance_cm\": 3.9}, {\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 8.5}]}", "target": "ball", "split": "train"}
{"id": "scene0388_00_0", "scene_id": "scene0388_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the broom appear?", "choices": null, "answer": "No", "target": "broom", "split": "train"}
{"id": "scene0388_00_1", "scene_id": "scene0388_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. office chair\nB. table\nC. chair\nD. monitor", "choices": ["A. office chair", "B. table", "C. chair", "D. monitor"], "answer": "A. office chair", "target": "backpack", "split": "train"}
{"id": "scene0390_00_0", "scene_id": "scene0390_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0390_00_1", "scene_id": "scene0390_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plate in this scene?\nA. chair\nB. monitor\nC. table\nD. plant", "choices": ["A. chair", "B. monitor", "C. table", "D. plant"], "answer": "D. plant", "target": "plate", "split": "train"}
{"id": "scene0390_00_2", "scene_id": "scene0390_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plate in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plate is on the table. It is also to the side of and below the plant, about 16 cm away. It is further to the side of and above the chair, about 29 cm away.", "target": "plate", "split": "train"}
{"id": "scene0390_00_3", "scene_id": "scene0390_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plate in structured JSON format.", "choices": null, "answer": "{\"target\": \"plate\", \"support_surface\": \"table\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 15.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 29.5}]}", "target": "plate", "split": "train"}
{"id": "scene0392_00_0", "scene_id": "scene0392_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0393_00_0", "scene_id": "scene0393_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0393_00_1", "scene_id": "scene0393_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. backpack\nC. chair\nD. desk", "choices": ["A. monitor", "B. backpack", "C. chair", "D. desk"], "answer": "C. chair", "target": "telephone", "split": "train"}
{"id": "scene0393_00_2", "scene_id": "scene0393_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the chair, about 44 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0393_00_3", "scene_id": "scene0393_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 44.0}]}", "target": "telephone", "split": "train"}
{"id": "scene0394_00_0", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0394_00_1", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. sign\nB. bin\nC. mirror\nD. plant", "choices": ["A. sign", "B. bin", "C. mirror", "D. plant"], "answer": "C. mirror", "target": "book", "split": "train"}
{"id": "scene0394_00_2", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the dresser. It is also to the side of and below the mirror, about 16 cm away. It is further to the side of and below the plant, about 26 cm away.", "target": "book", "split": "train"}
{"id": "scene0394_00_3", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 15.8}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 26.2}]}", "target": "book", "split": "train"}
{"id": "scene0394_00_4", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "Yes", "target": "water pitcher", "split": "train"}
{"id": "scene0394_00_5", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water pitcher in this scene?\nA. pillow\nB. sign\nC. plant\nD. bin", "choices": ["A. pillow", "B. sign", "C. plant", "D. bin"], "answer": "A. pillow", "target": "water pitcher", "split": "train"}
{"id": "scene0394_00_6", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water pitcher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water pitcher is on the nightstand. It is also to the side of and below the pillow, about 6 cm away. It is further next to the plant, about 8 cm away.", "target": "water pitcher", "split": "train"}
{"id": "scene0394_00_7", "scene_id": "scene0394_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water pitcher in structured JSON format.", "choices": null, "answer": "{\"target\": \"water pitcher\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 6.2}, {\"object\": \"plant\", \"relation\": \"next_to\", \"distance_cm\": 8.5}]}", "target": "water pitcher", "split": "train"}
{"id": "scene0395_00_0", "scene_id": "scene0395_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "scene0395_00_1", "scene_id": "scene0395_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the ball in this scene?\nA. poster tube\nB. desk\nC. bookshelf\nD. books", "choices": ["A. poster tube", "B. desk", "C. bookshelf", "D. books"], "answer": "D. books", "target": "ball", "split": "train"}
{"id": "scene0395_00_2", "scene_id": "scene0395_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the ball in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The ball is below the books, about 17 cm away. It is also to the side of and above the poster tube, about 34 cm away.", "target": "ball", "split": "train"}
{"id": "scene0395_00_3", "scene_id": "scene0395_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"books\", \"relation\": \"below\", \"distance_cm\": 17.2}, {\"object\": \"poster tube\", \"relation\": \"side_above\", \"distance_cm\": 34.5}]}", "target": "ball", "split": "train"}
{"id": "scene0396_00_0", "scene_id": "scene0396_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0396_00_1", "scene_id": "scene0396_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bathroom vanity\nB. sink\nC. toilet\nD. monitor", "choices": ["A. bathroom vanity", "B. sink", "C. toilet", "D. monitor"], "answer": "B. sink", "target": "toilet paper", "split": "train"}
{"id": "scene0396_00_2", "scene_id": "scene0396_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the sink, about 42 cm away. It is also to the side of and above the toilet, about 57 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0396_00_3", "scene_id": "scene0396_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom vanity\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 41.8}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 56.6}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0398_00_0", "scene_id": "scene0398_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "No", "target": "book", "split": "train"}
{"id": "scene0398_00_1", "scene_id": "scene0398_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. shelf\nB. monitor\nC. chair\nD. radiator", "choices": ["A. shelf", "B. monitor", "C. chair", "D. radiator"], "answer": "D. radiator", "target": "telephone", "split": "train"}
{"id": "scene0398_00_2", "scene_id": "scene0398_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the radiator, about 33 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0398_00_3", "scene_id": "scene0398_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 33.0}]}", "target": "telephone", "split": "train"}
{"id": "scene0399_00_0", "scene_id": "scene0399_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0400_00_0", "scene_id": "scene0400_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0400_00_1", "scene_id": "scene0400_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. chair\nC. recycling bin\nD. shelf", "choices": ["A. monitor", "B. chair", "C. recycling bin", "D. shelf"], "answer": "C. recycling bin", "target": "telephone", "split": "train"}
{"id": "scene0400_00_2", "scene_id": "scene0400_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the recycling bin, about 57 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0400_00_3", "scene_id": "scene0400_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"recycling bin\", \"relation\": \"side_above\", \"distance_cm\": 57.2}]}", "target": "telephone", "split": "train"}
{"id": "scene0403_00_0", "scene_id": "scene0403_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0403_00_1", "scene_id": "scene0403_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. monitor\nC. desk\nD. keyboard", "choices": ["A. chair", "B. monitor", "C. desk", "D. keyboard"], "answer": "A. chair", "target": "backpack", "split": "train"}
{"id": "scene0406_00_0", "scene_id": "scene0406_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "train"}
{"id": "scene0406_00_1", "scene_id": "scene0406_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. toilet\nB. trash can\nC. counter\nD. towel", "choices": ["A. toilet", "B. trash can", "C. counter", "D. towel"], "answer": "B. trash can", "target": "toilet paper", "split": "train"}
{"id": "scene0406_00_2", "scene_id": "scene0406_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the trash can, about 14 cm away. It is also to the side of and above the toilet, about 17 cm away. It is further to the side of and below the counter, about 28 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0406_00_3", "scene_id": "scene0406_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 14.1}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 17.4}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0410_00_0", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dustpan appear?", "choices": null, "answer": "No", "target": "dustpan", "split": "train"}
{"id": "scene0410_00_1", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. bathtub\nB. light switch\nC. bar\nD. towel", "choices": ["A. bathtub", "B. light switch", "C. bar", "D. towel"], "answer": "A. bathtub", "target": "soap dish", "split": "train"}
{"id": "scene0410_00_2", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the bathtub, about 11 cm away. It is also below the bar, about 13 cm away. It is further to the side of and above the towel, about 58 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0410_00_3", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"shower walls\", \"references\": [{\"object\": \"bathtub\", \"relation\": \"side_above\", \"distance_cm\": 11.1}, {\"object\": \"bar\", \"relation\": \"below\", \"distance_cm\": 12.7}]}", "target": "soap dish", "split": "train"}
{"id": "scene0410_00_4", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0410_00_5", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. light switch\nB. picture\nC. bathtub\nD. bar", "choices": ["A. light switch", "B. picture", "C. bathtub", "D. bar"], "answer": "A. light switch", "target": "toilet paper", "split": "train"}
{"id": "scene0410_00_6", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the light switch, about 54 cm away. It is also to the side of and below the picture, about 64 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0410_00_7", "scene_id": "scene0410_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"light switch\", \"references\": [{\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 53.9}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 64.3}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0413_00_0", "scene_id": "scene0413_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0413_00_1", "scene_id": "scene0413_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the shoe in this scene?\nA. shoe rack\nB. shoes\nC. closet\nD. clothing", "choices": ["A. shoe rack", "B. shoes", "C. closet", "D. clothing"], "answer": "C. closet", "target": "shoe", "split": "train"}
{"id": "scene0413_00_2", "scene_id": "scene0413_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the shoe in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The shoe is under the closet, about 3 cm away. It is also to the side of and below the clothing, about 30 cm away.", "target": "shoe", "split": "train"}
{"id": "scene0413_00_3", "scene_id": "scene0413_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the shoe in structured JSON format.", "choices": null, "answer": "{\"target\": \"shoe\", \"support_surface\": \"shoe rack\", \"references\": [{\"object\": \"closet\", \"relation\": \"under\", \"distance_cm\": 2.8}, {\"object\": \"clothing\", \"relation\": \"side_below\", \"distance_cm\": 29.7}]}", "target": "shoe", "split": "train"}
{"id": "scene0414_00_0", "scene_id": "scene0414_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0414_00_1", "scene_id": "scene0414_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. desk\nC. keyboard\nD. chair", "choices": ["A. monitor", "B. desk", "C. keyboard", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "train"}
{"id": "scene0415_00_0", "scene_id": "scene0415_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0415_00_1", "scene_id": "scene0415_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. trash can\nB. towel\nC. toilet\nD. bathroom vanity", "choices": ["A. trash can", "B. towel", "C. toilet", "D. bathroom vanity"], "answer": "A. trash can", "target": "toilet paper", "split": "train"}
{"id": "scene0415_00_2", "scene_id": "scene0415_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is above the trash can, about 8 cm away. It is also near the toilet, about 17 cm away. It is further to the side of and below the bathroom vanity, about 30 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0415_00_3", "scene_id": "scene0415_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"trash can\", \"relation\": \"above\", \"distance_cm\": 7.8}, {\"object\": \"toilet\", \"relation\": \"near\", \"distance_cm\": 16.9}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0416_00_0", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "scene0416_00_1", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. purse\nC. clothing\nD. laptop", "choices": ["A. bag", "B. purse", "C. clothing", "D. laptop"], "answer": "C. clothing", "target": "backpack", "split": "train"}
{"id": "scene0416_00_2", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the clothing, about 33 cm away. It is also to the side of and below the purse, about 40 cm away. It is further to the side of and below the laptop, about 60 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0416_00_3", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"shoes\", \"references\": [{\"object\": \"clothing\", \"relation\": \"side_below\", \"distance_cm\": 33.3}, {\"object\": \"purse\", \"relation\": \"side_below\", \"distance_cm\": 39.7}]}", "target": "backpack", "split": "train"}
{"id": "scene0416_00_4", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "train"}
{"id": "scene0416_00_5", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. bed\nB. case of water bottles\nC. suitcase\nD. backpack", "choices": ["A. bed", "B. case of water bottles", "C. suitcase", "D. backpack"], "answer": "C. suitcase", "target": "bag", "split": "train"}
{"id": "scene0416_00_6", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the suitcase, about 20 cm away. It is also to the side of and above the case of water bottles, about 38 cm away.", "target": "bag", "split": "train"}
{"id": "scene0416_00_7", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"suitcase\", \"references\": [{\"object\": \"suitcase\", \"relation\": \"side_above\", \"distance_cm\": 19.9}, {\"object\": \"case of water bottles\", \"relation\": \"side_above\", \"distance_cm\": 37.7}]}", "target": "bag", "split": "train"}
{"id": "scene0416_00_8", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0416_00_9", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. pillow\nB. backpack\nC. purse\nD. clothing", "choices": ["A. pillow", "B. backpack", "C. purse", "D. clothing"], "answer": "C. purse", "target": "laptop", "split": "train"}
{"id": "scene0416_00_10", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the bed. It is also to the side of and above the purse, about 21 cm away. It is further to the side of and above the clothing, about 26 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0416_00_11", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"purse\", \"relation\": \"side_above\", \"distance_cm\": 21.4}, {\"object\": \"clothing\", \"relation\": \"side_above\", \"distance_cm\": 26.4}]}", "target": "laptop", "split": "train"}
{"id": "scene0416_00_12", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0416_00_13", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the purse in this scene?\nA. laptop\nB. shoes\nC. bag\nD. backpack", "choices": ["A. laptop", "B. shoes", "C. bag", "D. backpack"], "answer": "A. laptop", "target": "purse", "split": "train"}
{"id": "scene0416_00_14", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the purse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The purse is to the side of and below the laptop, about 21 cm away. It is also to the side of and above the backpack, about 40 cm away. It is further to the side of and above the shoes, about 77 cm away.", "target": "purse", "split": "train"}
{"id": "scene0416_00_15", "scene_id": "scene0416_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the purse in structured JSON format.", "choices": null, "answer": "{\"target\": \"purse\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 21.4}, {\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 39.7}]}", "target": "purse", "split": "train"}
{"id": "scene0417_00_0", "scene_id": "scene0417_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0419_00_0", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0419_00_1", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. keyboard\nB. chair\nC. pillow\nD. desk", "choices": ["A. keyboard", "B. chair", "C. pillow", "D. desk"], "answer": "C. pillow", "target": "backpack", "split": "train"}
{"id": "scene0419_00_2", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the pillow, about 18 cm away. It is also to the side of and below the chair, about 38 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0419_00_3", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 18.3}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 38.5}]}", "target": "backpack", "split": "train"}
{"id": "scene0419_00_4", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0419_00_5", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. monitor\nC. backpack\nD. laptop", "choices": ["A. chair", "B. monitor", "C. backpack", "D. laptop"], "answer": "D. laptop", "target": "keyboard", "split": "train"}
{"id": "scene0419_00_6", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the laptop, about 13 cm away. It is also to the side of and below the monitor, about 18 cm away. It is further to the side of and above the chair, about 28 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0419_00_7", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 12.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.1}]}", "target": "keyboard", "split": "train"}
{"id": "scene0419_00_8", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0419_00_9", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. keyboard\nB. monitor\nC. pillow\nD. mouse", "choices": ["A. keyboard", "B. monitor", "C. pillow", "D. mouse"], "answer": "B. monitor", "target": "laptop", "split": "train"}
{"id": "scene0419_00_10", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also to the side of and below the monitor, about 5 cm away. It is further to the side of and above the keyboard, about 13 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0419_00_11", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.2}, {\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 12.7}]}", "target": "laptop", "split": "train"}
{"id": "scene0419_00_12", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "scene0419_00_13", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. pillow\nC. laptop\nD. monitor", "choices": ["A. chair", "B. pillow", "C. laptop", "D. monitor"], "answer": "C. laptop", "target": "mouse", "split": "train"}
{"id": "scene0419_00_14", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and below the laptop, about 13 cm away. It is also to the side of and below the monitor, about 30 cm away. It is further to the side of and above the chair, about 55 cm away.", "target": "mouse", "split": "train"}
{"id": "scene0419_00_15", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 12.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 29.6}]}", "target": "mouse", "split": "train"}
{"id": "scene0419_00_16", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0419_00_17", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper towel roll in this scene?\nA. kitchen cabinets\nB. kitchen cabinet\nC. backpack\nD. towel", "choices": ["A. kitchen cabinets", "B. kitchen cabinet", "C. backpack", "D. towel"], "answer": "A. kitchen cabinets", "target": "paper towel roll", "split": "train"}
{"id": "scene0419_00_18", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper towel roll in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper towel roll is to the side of and below the kitchen cabinets, about 21 cm away. It is also to the side of and above the kitchen cabinet, about 27 cm away. It is further to the side of and above the towel, about 33 cm away.", "target": "paper towel roll", "split": "train"}
{"id": "scene0419_00_19", "scene_id": "scene0419_00", "source_dataset": "scannet200", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen cabinets\", \"relation\": \"side_below\", \"distance_cm\": 21.4}, {\"object\": \"kitchen cabinet\", \"relation\": \"side_above\", \"distance_cm\": 27.4}]}", "target": "paper towel roll", "split": "train"}
{"id": "scene0421_00_0", "scene_id": "scene0421_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "scene0421_00_1", "scene_id": "scene0421_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the broom in this scene?\nA. monitor\nB. basket\nC. cloth\nD. box", "choices": ["A. monitor", "B. basket", "C. cloth", "D. box"], "answer": "C. cloth", "target": "broom", "split": "train"}
{"id": "scene0421_00_2", "scene_id": "scene0421_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the broom in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The broom is to the side of and below the cloth, about 7 cm away. It is also to the side of and above the box, about 13 cm away. It is further to the side of and below the basket, about 53 cm away.", "target": "broom", "split": "train"}
{"id": "scene0421_00_3", "scene_id": "scene0421_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the broom in structured JSON format.", "choices": null, "answer": "{\"target\": \"broom\", \"support_surface\": \"box\", \"references\": [{\"object\": \"cloth\", \"relation\": \"side_below\", \"distance_cm\": 7.3}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 13.2}]}", "target": "broom", "split": "train"}
{"id": "scene0424_00_0", "scene_id": "scene0424_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0424_00_1", "scene_id": "scene0424_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"couch\", \"references\": []}", "target": "backpack", "split": "train"}
{"id": "scene0424_00_2", "scene_id": "scene0424_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0424_00_3", "scene_id": "scene0424_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the dumbbell in this scene?\nA. shelf\nB. chair\nC. backpack\nD. book", "choices": ["A. shelf", "B. chair", "C. backpack", "D. book"], "answer": "B. chair", "target": "dumbbell", "split": "train"}
{"id": "scene0424_00_4", "scene_id": "scene0424_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the dumbbell in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The dumbbell is to the side of and below the shelf, about 43 cm away.", "target": "dumbbell", "split": "train"}
{"id": "scene0424_00_5", "scene_id": "scene0424_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the dumbbell in structured JSON format.", "choices": null, "answer": "{\"target\": \"dumbbell\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 43.2}]}", "target": "dumbbell", "split": "train"}
{"id": "scene0425_00_0", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0425_00_1", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. recycling bin\nC. chair\nD. stool", "choices": ["A. desk", "B. recycling bin", "C. chair", "D. stool"], "answer": "D. stool", "target": "backpack", "split": "train"}
{"id": "scene0425_00_2", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the stool. It is also to the side of and below the chair, about 19 cm away. It is further near the recycling bin, about 24 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0425_00_3", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"stool\", \"relation\": \"side_above\", \"distance_cm\": 2.2}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 18.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0425_00_4", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "No", "target": "bottle", "split": "train"}
{"id": "scene0425_00_5", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. chair\nC. dresser\nD. wall hanging", "choices": ["A. monitor", "B. chair", "C. dresser", "D. wall hanging"], "answer": "A. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0425_00_6", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and below the monitor, about 19 cm away. It is further to the side of and above the dresser, about 24 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0425_00_7", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 19.2}, {\"object\": \"dresser\", \"relation\": \"side_above\", \"distance_cm\": 24.1}]}", "target": "keyboard", "split": "train"}
{"id": "scene0425_00_8", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0425_00_9", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water pitcher in this scene?\nA. backpack\nB. laundry hamper\nC. heater\nD. chair", "choices": ["A. backpack", "B. laundry hamper", "C. heater", "D. chair"], "answer": "B. laundry hamper", "target": "water pitcher", "split": "train"}
{"id": "scene0425_00_10", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water pitcher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water pitcher is on the shelf. It is also to the side of and above the laundry hamper, about 27 cm away. It is further to the side of and above the heater, about 47 cm away.", "target": "water pitcher", "split": "train"}
{"id": "scene0425_00_11", "scene_id": "scene0425_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water pitcher in structured JSON format.", "choices": null, "answer": "{\"target\": \"water pitcher\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"laundry hamper\", \"relation\": \"side_above\", \"distance_cm\": 27.0}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 46.7}]}", "target": "water pitcher", "split": "train"}
{"id": "scene0426_00_0", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0426_00_1", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. monitor\nC. jacket\nD. box", "choices": ["A. chair", "B. monitor", "C. jacket", "D. box"], "answer": "C. jacket", "target": "backpack", "split": "train"}
{"id": "scene0426_00_2", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the jacket, about 8 cm away. It is also to the side of and below the box, about 11 cm away. It is further near the monitor, about 49 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0426_00_3", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"organizer shelf\", \"references\": [{\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 7.6}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 10.5}]}", "target": "backpack", "split": "train"}
{"id": "scene0426_00_4", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "scene0426_00_5", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. closet doors\nB. chair\nC. box\nD. lamp", "choices": ["A. closet doors", "B. chair", "C. box", "D. lamp"], "answer": "A. closet doors", "target": "tissue box", "split": "train"}
{"id": "scene0426_00_6", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and below the closet doors, about 31 cm away. It is also to the side of and below the lamp, about 44 cm away. It is further to the side of and above the box, about 67 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0426_00_7", "scene_id": "scene0426_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"closet doors\", \"relation\": \"side_below\", \"distance_cm\": 31.0}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 44.5}]}", "target": "tissue box", "split": "train"}
{"id": "scene0429_00_0", "scene_id": "scene0429_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the fire extinguisher appear?", "choices": null, "answer": "No", "target": "fire extinguisher", "split": "train"}
{"id": "scene0429_00_1", "scene_id": "scene0429_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. trash can\nB. toilet\nC. monitor\nD. chair", "choices": ["A. trash can", "B. toilet", "C. monitor", "D. chair"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0429_00_2", "scene_id": "scene0429_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the toilet, about 14 cm away. It is also near the trash can, about 47 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0429_00_3", "scene_id": "scene0429_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 14.4}, {\"object\": \"trash can\", \"relation\": \"near\", \"distance_cm\": 46.9}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0431_00_0", "scene_id": "scene0431_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0433_00_0", "scene_id": "scene0433_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0433_00_1", "scene_id": "scene0433_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. chair\nC. dining table\nD. monitor", "choices": ["A. desk", "B. chair", "C. dining table", "D. monitor"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0435_00_0", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper towel roll appear?", "choices": null, "answer": "No", "target": "paper towel roll", "split": "train"}
{"id": "scene0435_00_1", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothing\nB. sink\nC. pillow\nD. bathroom vanity", "choices": ["A. clothing", "B. sink", "C. pillow", "D. bathroom vanity"], "answer": "A. clothing", "target": "backpack", "split": "train"}
{"id": "scene0435_00_2", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the clothing, about 39 cm away. It is also to the side of and below the pillow, about 56 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0435_00_3", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"sofa chair\", \"references\": [{\"object\": \"clothing\", \"relation\": \"side_below\", \"distance_cm\": 39.4}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 55.9}]}", "target": "backpack", "split": "train"}
{"id": "scene0435_00_4", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0435_00_5", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. toilet\nB. trash can\nC. towel\nD. bathroom vanity", "choices": ["A. toilet", "B. trash can", "C. towel", "D. bathroom vanity"], "answer": "B. trash can", "target": "toilet paper", "split": "train"}
{"id": "scene0435_00_6", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the trash can, about 13 cm away. It is also to the side of and above the toilet, about 17 cm away. It is further to the side of and below the bathroom vanity, about 28 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0435_00_7", "scene_id": "scene0435_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"trash can\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 13.1}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 17.0}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0436_00_0", "scene_id": "scene0436_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0436_00_1", "scene_id": "scene0436_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. kitchen counter\nB. kitchen cabinets\nC. sink\nD. microwave", "choices": ["A. kitchen counter", "B. kitchen cabinets", "C. sink", "D. microwave"], "answer": "A. kitchen counter", "target": "soap dispenser", "split": "train"}
{"id": "scene0436_00_2", "scene_id": "scene0436_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the kitchen counter, about 6 cm away. It is also to the side of and above the microwave, about 15 cm away. It is further to the side of and above the sink, about 16 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0436_00_3", "scene_id": "scene0436_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 5.9}, {\"object\": \"microwave\", \"relation\": \"side_above\", \"distance_cm\": 15.3}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0439_00_0", "scene_id": "scene0439_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0439_00_1", "scene_id": "scene0439_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. kitchen cabinets\nB. dish rack\nC. trash can\nD. recycling bin", "choices": ["A. kitchen cabinets", "B. dish rack", "C. trash can", "D. recycling bin"], "answer": "D. recycling bin", "target": "paper bag", "split": "train"}
{"id": "scene0439_00_2", "scene_id": "scene0439_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is next to the recycling bin, about 3 cm away. It is also to the side of and below the trash can, about 14 cm away. It is further to the side of and below the dish rack, about 66 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0439_00_3", "scene_id": "scene0439_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"dish rack\", \"references\": [{\"object\": \"recycling bin\", \"relation\": \"next_to\", \"distance_cm\": 3.1}, {\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 13.6}]}", "target": "paper bag", "split": "train"}
{"id": "scene0440_00_0", "scene_id": "scene0440_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0440_00_1", "scene_id": "scene0440_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. office chair\nB. shelf\nC. desk\nD. monitor", "choices": ["A. office chair", "B. shelf", "C. desk", "D. monitor"], "answer": "C. desk", "target": "telephone", "split": "train"}
{"id": "scene0440_00_2", "scene_id": "scene0440_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the desk, about 24 cm away. It is also to the side of and below the shelf, about 43 cm away. It is further to the side of and above the office chair, about 52 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0440_00_3", "scene_id": "scene0440_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 23.7}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 43.3}]}", "target": "telephone", "split": "train"}
{"id": "scene0443_00_0", "scene_id": "scene0443_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0443_00_1", "scene_id": "scene0443_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. computer tower\nB. board\nC. monitor\nD. chair", "choices": ["A. computer tower", "B. board", "C. monitor", "D. chair"], "answer": "C. monitor", "target": "laptop", "split": "train"}
{"id": "scene0443_00_2", "scene_id": "scene0443_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also to the side of and below the monitor, about 25 cm away. It is further to the side of and above the chair, about 35 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0443_00_3", "scene_id": "scene0443_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 24.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 34.7}]}", "target": "laptop", "split": "train"}
{"id": "scene0447_00_0", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0447_00_1", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plunger in this scene?\nA. toilet paper\nB. trash can\nC. toilet\nD. shower", "choices": ["A. toilet paper", "B. trash can", "C. toilet", "D. shower"], "answer": "C. toilet", "target": "plunger", "split": "train"}
{"id": "scene0447_00_2", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plunger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plunger is to the side of and below the toilet, about 9 cm away. It is also to the side of and below the shower, about 18 cm away. It is further near the trash can, about 31 cm away.", "target": "plunger", "split": "train"}
{"id": "scene0447_00_3", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plunger in structured JSON format.", "choices": null, "answer": "{\"target\": \"plunger\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 9.1}, {\"object\": \"shower\", \"relation\": \"side_below\", \"distance_cm\": 18.2}]}", "target": "plunger", "split": "train"}
{"id": "scene0447_00_4", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "Yes", "target": "soap dish", "split": "train"}
{"id": "scene0447_00_5", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. plunger\nB. clothes\nC. bathrobe\nD. bathtub", "choices": ["A. plunger", "B. clothes", "C. bathrobe", "D. bathtub"], "answer": "D. bathtub", "target": "soap dish", "split": "train"}
{"id": "scene0447_00_6", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the bathtub, about 20 cm away. It is also to the side of and below the clothes, about 30 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0447_00_7", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"bathtub\", \"references\": [{\"object\": \"bathtub\", \"relation\": \"side_above\", \"distance_cm\": 20.1}, {\"object\": \"clothes\", \"relation\": \"side_below\", \"distance_cm\": 30.5}]}", "target": "soap dish", "split": "train"}
{"id": "scene0447_00_8", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the broom appear?", "choices": null, "answer": "No", "target": "broom", "split": "train"}
{"id": "scene0447_00_9", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. plunger\nB. toilet\nC. bathrobe\nD. trash can", "choices": ["A. plunger", "B. toilet", "C. bathrobe", "D. trash can"], "answer": "B. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0447_00_10", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 25 cm away. It is also to the side of and above the trash can, about 39 cm away. It is further to the side of and above the plunger, about 57 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0447_00_11", "scene_id": "scene0447_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"shower\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 25.0}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 39.0}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0449_00_0", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0449_00_1", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. mirror\nB. bathroom vanity\nC. bar\nD. toilet paper", "choices": ["A. mirror", "B. bathroom vanity", "C. bar", "D. toilet paper"], "answer": "D. toilet paper", "target": "bag", "split": "train"}
{"id": "scene0449_00_2", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the toilet paper, about 84 cm away.", "target": "bag", "split": "train"}
{"id": "scene0449_00_3", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"side_above\", \"distance_cm\": 84.3}]}", "target": "bag", "split": "train"}
{"id": "scene0449_00_4", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0449_00_5", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. towel\nB. sink\nC. toilet paper\nD. mirror", "choices": ["A. towel", "B. sink", "C. toilet paper", "D. mirror"], "answer": "B. sink", "target": "soap dish", "split": "train"}
{"id": "scene0449_00_6", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the sink, about 6 cm away. It is also to the side of and below the mirror, about 25 cm away. It is further to the side of and below the towel, about 68 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0449_00_7", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 25.0}]}", "target": "soap dish", "split": "train"}
{"id": "scene0449_00_8", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "scene0449_00_9", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. shower walls\nB. toilet\nC. towel\nD. toilet paper", "choices": ["A. shower walls", "B. toilet", "C. towel", "D. toilet paper"], "answer": "A. shower walls", "target": "tissue box", "split": "train"}
{"id": "scene0449_00_10", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and below the shower walls, about 13 cm away. It is also above the toilet paper, about 16 cm away. It is further to the side of and above the towel, about 20 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0449_00_11", "scene_id": "scene0449_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"shower walls\", \"relation\": \"side_below\", \"distance_cm\": 13.0}, {\"object\": \"toilet paper\", \"relation\": \"above\", \"distance_cm\": 16.1}]}", "target": "tissue box", "split": "train"}
{"id": "scene0450_00_0", "scene_id": "scene0450_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0450_00_1", "scene_id": "scene0450_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. desk\nC. monitor\nD. sofa chair", "choices": ["A. chair", "B. desk", "C. monitor", "D. sofa chair"], "answer": "D. sofa chair", "target": "backpack", "split": "train"}
{"id": "scene0450_00_2", "scene_id": "scene0450_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the sofa chair, about 34 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0450_00_3", "scene_id": "scene0450_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"sofa chair\", \"relation\": \"side_above\", \"distance_cm\": 34.1}]}", "target": "backpack", "split": "train"}
{"id": "scene0451_00_0", "scene_id": "scene0451_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0451_00_1", "scene_id": "scene0451_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. cabinet\nB. chair\nC. plant\nD. monitor", "choices": ["A. cabinet", "B. chair", "C. plant", "D. monitor"], "answer": "C. plant", "target": "bag", "split": "train"}
{"id": "scene0451_00_2", "scene_id": "scene0451_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the plant, about 27 cm away.", "target": "bag", "split": "train"}
{"id": "scene0451_00_3", "scene_id": "scene0451_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 26.9}]}", "target": "bag", "split": "train"}
{"id": "scene0452_00_0", "scene_id": "scene0452_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0455_00_0", "scene_id": "scene0455_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hat appear?", "choices": null, "answer": "No", "target": "hat", "split": "train"}
{"id": "scene0455_00_1", "scene_id": "scene0455_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. chair\nC. box\nD. desk", "choices": ["A. monitor", "B. chair", "C. box", "D. desk"], "answer": "C. box", "target": "backpack", "split": "train"}
{"id": "scene0455_00_2", "scene_id": "scene0455_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the box, about 32 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0455_00_3", "scene_id": "scene0455_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 31.6}]}", "target": "backpack", "split": "train"}
{"id": "scene0456_00_0", "scene_id": "scene0456_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0456_00_1", "scene_id": "scene0456_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. office chair\nB. tv\nC. table\nD. monitor", "choices": ["A. office chair", "B. tv", "C. table", "D. monitor"], "answer": "B. tv", "target": "backpack", "split": "train"}
{"id": "scene0456_00_2", "scene_id": "scene0456_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and below the tv, about 50 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0456_00_3", "scene_id": "scene0456_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 49.9}]}", "target": "backpack", "split": "train"}
{"id": "scene0457_00_0", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0457_00_1", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. cabinet\nB. desk\nC. tissue box\nD. office chair", "choices": ["A. cabinet", "B. desk", "C. tissue box", "D. office chair"], "answer": "C. tissue box", "target": "book", "split": "train"}
{"id": "scene0457_00_2", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is to the side of and below the tissue box, about 9 cm away. It is also near the office chair, about 60 cm away.", "target": "book", "split": "train"}
{"id": "scene0457_00_3", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"tissue box\", \"relation\": \"side_below\", \"distance_cm\": 9.2}, {\"object\": \"office chair\", \"relation\": \"near\", \"distance_cm\": 59.9}]}", "target": "book", "split": "train"}
{"id": "scene0457_00_4", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0457_00_5", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scale in this scene?\nA. sofa chair\nB. cabinet\nC. book\nD. desk", "choices": ["A. sofa chair", "B. cabinet", "C. book", "D. desk"], "answer": "A. sofa chair", "target": "scale", "split": "train"}
{"id": "scene0457_00_6", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is to the side of and below the sofa chair, about 23 cm away. It is also to the side of and below the cabinet, about 66 cm away.", "target": "scale", "split": "train"}
{"id": "scene0457_00_7", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"sofa chair\", \"relation\": \"side_below\", \"distance_cm\": 22.8}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 65.8}]}", "target": "scale", "split": "train"}
{"id": "scene0457_00_8", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "scene0457_00_9", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. cabinet\nB. office chair\nC. desk\nD. book", "choices": ["A. cabinet", "B. office chair", "C. desk", "D. book"], "answer": "D. book", "target": "tissue box", "split": "train"}
{"id": "scene0457_00_10", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and above the book, about 9 cm away. It is also to the side of and above the office chair, about 62 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0457_00_11", "scene_id": "scene0457_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 9.2}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 61.7}]}", "target": "tissue box", "split": "train"}
{"id": "scene0458_00_0", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the broom appear?", "choices": null, "answer": "Yes", "target": "broom", "split": "train"}
{"id": "scene0458_00_1", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the broom in this scene?\nA. toilet paper\nB. toilet\nC. trash can\nD. bottle", "choices": ["A. toilet paper", "B. toilet", "C. trash can", "D. bottle"], "answer": "A. toilet paper", "target": "broom", "split": "train"}
{"id": "scene0458_00_2", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the broom in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The broom is under the toilet paper, about 3 cm away. It is also to the side of and above the bottle, about 10 cm away. It is further to the side of and below the toilet, about 26 cm away.", "target": "broom", "split": "train"}
{"id": "scene0458_00_3", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the broom in structured JSON format.", "choices": null, "answer": "{\"target\": \"broom\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"under\", \"distance_cm\": 2.8}, {\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 10.0}]}", "target": "broom", "split": "train"}
{"id": "scene0458_00_4", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0458_00_5", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the dustpan in this scene?\nA. bottle\nB. trash can\nC. toilet paper\nD. toilet", "choices": ["A. bottle", "B. trash can", "C. toilet paper", "D. toilet"], "answer": "A. bottle", "target": "dustpan", "split": "train"}
{"id": "scene0458_00_6", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the dustpan in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The dustpan is to the side of and above the bottle, about 5 cm away. It is also to the side of and below the toilet, about 13 cm away. It is further near the trash can, about 58 cm away.", "target": "dustpan", "split": "train"}
{"id": "scene0458_00_7", "scene_id": "scene0458_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the dustpan in structured JSON format.", "choices": null, "answer": "{\"target\": \"dustpan\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 5.0}, {\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 12.6}]}", "target": "dustpan", "split": "train"}
{"id": "scene0460_00_0", "scene_id": "scene0460_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0460_00_1", "scene_id": "scene0460_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. toilet\nB. bathroom vanity\nC. monitor\nD. sink", "choices": ["A. toilet", "B. bathroom vanity", "C. monitor", "D. sink"], "answer": "A. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0460_00_2", "scene_id": "scene0460_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 16 cm away. It is also to the side of and below the sink, about 32 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0460_00_3", "scene_id": "scene0460_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"toilet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 15.7}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 31.9}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0464_00_0", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0464_00_1", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. books\nB. box\nC. calendar\nD. chair", "choices": ["A. books", "B. box", "C. calendar", "D. chair"], "answer": "A. books", "target": "book", "split": "train"}
{"id": "scene0464_00_2", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is to the side of and below the books, about 7 cm away. It is also to the side of and above the box, about 30 cm away. It is further to the side of and above the calendar, about 56 cm away.", "target": "book", "split": "train"}
{"id": "scene0464_00_3", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"books\", \"references\": [{\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 6.8}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 29.8}]}", "target": "book", "split": "train"}
{"id": "scene0464_00_4", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "scene0464_00_5", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. paper\nB. keyboard\nC. cup\nD. chair", "choices": ["A. paper", "B. keyboard", "C. cup", "D. chair"], "answer": "C. cup", "target": "headphones", "split": "train"}
{"id": "scene0464_00_6", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the desk. It is also next to the cup, about 3 cm away. It is further to the side of and below the chair, about 11 cm away.", "target": "headphones", "split": "train"}
{"id": "scene0464_00_7", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 10.6}]}", "target": "headphones", "split": "train"}
{"id": "scene0464_00_8", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "No", "target": "telephone", "split": "train"}
{"id": "scene0464_00_9", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. sweater\nB. book\nC. books\nD. monitor", "choices": ["A. sweater", "B. book", "C. books", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0464_00_10", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the monitor, about 17 cm away. It is also to the side of and below the books, about 26 cm away. It is further to the side of and above the sweater, about 26 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0464_00_11", "scene_id": "scene0464_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"books\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 17.3}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 25.6}]}", "target": "keyboard", "split": "train"}
{"id": "scene0465_00_0", "scene_id": "scene0465_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0465_00_1", "scene_id": "scene0465_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. monitor\nB. pantry wall\nC. chair\nD. desk", "choices": ["A. monitor", "B. pantry wall", "C. chair", "D. desk"], "answer": "B. pantry wall", "target": "paper bag", "split": "train"}
{"id": "scene0465_00_2", "scene_id": "scene0465_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is to the side of and below the pantry wall, about 43 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0465_00_3", "scene_id": "scene0465_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"pantry wall\", \"references\": [{\"object\": \"pantry wall\", \"relation\": \"side_below\", \"distance_cm\": 42.6}]}", "target": "paper bag", "split": "train"}
{"id": "scene0467_00_0", "scene_id": "scene0467_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0467_00_1", "scene_id": "scene0467_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. table\nB. chair\nC. pillow\nD. monitor", "choices": ["A. table", "B. chair", "C. pillow", "D. monitor"], "answer": "C. pillow", "target": "book", "split": "train"}
{"id": "scene0467_00_2", "scene_id": "scene0467_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the table. It is also near the pillow, about 23 cm away.", "target": "book", "split": "train"}
{"id": "scene0467_00_3", "scene_id": "scene0467_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pillow\", \"relation\": \"near\", \"distance_cm\": 23.0}]}", "target": "book", "split": "train"}
{"id": "scene0468_00_0", "scene_id": "scene0468_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "No", "target": "book", "split": "train"}
{"id": "scene0468_00_1", "scene_id": "scene0468_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. monitor\nB. bookshelf\nC. shelf\nD. office chair", "choices": ["A. monitor", "B. bookshelf", "C. shelf", "D. office chair"], "answer": "B. bookshelf", "target": "telephone", "split": "train"}
{"id": "scene0468_00_2", "scene_id": "scene0468_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the bookshelf, about 25 cm away. It is also to the side of and above the office chair, about 35 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0468_00_3", "scene_id": "scene0468_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"side_above\", \"distance_cm\": 25.3}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 35.1}]}", "target": "telephone", "split": "train"}
{"id": "scene0469_00_0", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0469_00_1", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. computer tower\nB. monitor\nC. book\nD. chair", "choices": ["A. computer tower", "B. monitor", "C. book", "D. chair"], "answer": "B. monitor", "target": "bag", "split": "train"}
{"id": "scene0469_00_2", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the monitor, about 20 cm away. It is also to the side of and below the book, about 23 cm away. It is further to the side of and above the computer tower, about 35 cm away.", "target": "bag", "split": "train"}
{"id": "scene0469_00_3", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 20.2}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 23.1}]}", "target": "bag", "split": "train"}
{"id": "scene0469_00_4", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0469_00_5", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. bag\nB. keyboard\nC. monitor\nD. chair", "choices": ["A. bag", "B. keyboard", "C. monitor", "D. chair"], "answer": "B. keyboard", "target": "book", "split": "train"}
{"id": "scene0469_00_6", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the desk. It is also next to the keyboard, about 14 cm away. It is further to the side of and below the monitor, about 18 cm away.", "target": "book", "split": "train"}
{"id": "scene0469_00_7", "scene_id": "scene0469_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 14.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 17.6}]}", "target": "book", "split": "train"}
{"id": "scene0470_00_0", "scene_id": "scene0470_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0470_00_1", "scene_id": "scene0470_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. chair\nB. monitor\nC. desk\nD. backpack", "choices": ["A. chair", "B. monitor", "C. desk", "D. backpack"], "answer": "A. chair", "target": "bag", "split": "train"}
{"id": "scene0473_00_0", "scene_id": "scene0473_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0473_00_1", "scene_id": "scene0473_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dustpan appear?", "choices": null, "answer": "No", "target": "dustpan", "split": "train"}
{"id": "scene0474_00_0", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0474_00_1", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. chair\nB. monitor\nC. backpack\nD. keyboard", "choices": ["A. chair", "B. monitor", "C. backpack", "D. keyboard"], "answer": "D. keyboard", "target": "laptop", "split": "train"}
{"id": "scene0474_00_2", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also near the keyboard, about 23 cm away. It is further to the side of and above the chair, about 47 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0474_00_3", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 22.8}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 47.3}]}", "target": "laptop", "split": "train"}
{"id": "scene0474_00_4", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "Yes", "target": "paper bag", "split": "train"}
{"id": "scene0474_00_5", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper bag in this scene?\nA. backpack\nB. computer tower\nC. chair\nD. clothing", "choices": ["A. backpack", "B. computer tower", "C. chair", "D. clothing"], "answer": "A. backpack", "target": "paper bag", "split": "train"}
{"id": "scene0474_00_6", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper bag is to the side of and above the backpack, about 8 cm away.", "target": "paper bag", "split": "train"}
{"id": "scene0474_00_7", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"backpack\", \"references\": [{\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 7.5}]}", "target": "paper bag", "split": "train"}
{"id": "scene0474_00_8", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the broom appear?", "choices": null, "answer": "No", "target": "broom", "split": "train"}
{"id": "scene0474_00_9", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. person\nB. chair\nC. monitor\nD. computer tower", "choices": ["A. person", "B. chair", "C. monitor", "D. computer tower"], "answer": "C. monitor", "target": "telephone", "split": "train"}
{"id": "scene0474_00_10", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and below the monitor, about 8 cm away. It is also to the side of and above the computer tower, about 34 cm away. It is further to the side of and above the person, about 44 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0474_00_11", "scene_id": "scene0474_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.3}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 33.6}]}", "target": "telephone", "split": "train"}
{"id": "scene0478_00_0", "scene_id": "scene0478_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "No", "target": "book", "split": "train"}
{"id": "scene0478_00_1", "scene_id": "scene0478_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tray in this scene?\nA. copier\nB. chair\nC. monitor\nD. machine", "choices": ["A. copier", "B. chair", "C. monitor", "D. machine"], "answer": "D. machine", "target": "tray", "split": "train"}
{"id": "scene0478_00_2", "scene_id": "scene0478_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tray in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tray is under the machine, about 3 cm away.", "target": "tray", "split": "train"}
{"id": "scene0478_00_3", "scene_id": "scene0478_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tray in structured JSON format.", "choices": null, "answer": "{\"target\": \"tray\", \"support_surface\": \"copier\", \"references\": [{\"object\": \"machine\", \"relation\": \"under\", \"distance_cm\": 2.9}]}", "target": "tray", "split": "train"}
{"id": "scene0479_00_0", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0479_00_1", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. kitchen cabinets\nB. kitchen cabinet\nC. dishwasher\nD. dish rack", "choices": ["A. kitchen cabinets", "B. kitchen cabinet", "C. dishwasher", "D. dish rack"], "answer": "D. dish rack", "target": "cup", "split": "train"}
{"id": "scene0479_00_2", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the dish rack, about 10 cm away. It is also to the side of and above the dishwasher, about 22 cm away. It is further to the side of and above the kitchen cabinet, about 32 cm away.", "target": "cup", "split": "train"}
{"id": "scene0479_00_3", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"dish rack\", \"references\": [{\"object\": \"dish rack\", \"relation\": \"next_to\", \"distance_cm\": 9.5}, {\"object\": \"dishwasher\", \"relation\": \"side_above\", \"distance_cm\": 22.2}]}", "target": "cup", "split": "train"}
{"id": "scene0479_00_4", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "No", "target": "telephone", "split": "train"}
{"id": "scene0479_00_5", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the fire extinguisher in this scene?\nA. dish rack\nB. box\nC. bottle\nD. cup", "choices": ["A. dish rack", "B. box", "C. bottle", "D. cup"], "answer": "B. box", "target": "fire extinguisher", "split": "train"}
{"id": "scene0479_00_6", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the fire extinguisher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The fire extinguisher is to the side of and below the box, about 55 cm away. It is also to the side of and below the bottle, about 69 cm away.", "target": "fire extinguisher", "split": "train"}
{"id": "scene0479_00_7", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the fire extinguisher in structured JSON format.", "choices": null, "answer": "{\"target\": \"fire extinguisher\", \"support_surface\": \"box\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 54.7}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 69.2}]}", "target": "fire extinguisher", "split": "train"}
{"id": "scene0479_00_8", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "Yes", "target": "toaster oven", "split": "train"}
{"id": "scene0479_00_9", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toaster oven in this scene?\nA. kitchen cabinet\nB. envelope\nC. box\nD. bottle", "choices": ["A. kitchen cabinet", "B. envelope", "C. box", "D. bottle"], "answer": "A. kitchen cabinet", "target": "toaster oven", "split": "train"}
{"id": "scene0479_00_10", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toaster oven in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster oven is to the side of and above the kitchen cabinet, about 38 cm away. It is also to the side of and below the envelope, about 41 cm away.", "target": "toaster oven", "split": "train"}
{"id": "scene0479_00_11", "scene_id": "scene0479_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toaster oven in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster oven\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"kitchen cabinet\", \"relation\": \"side_above\", \"distance_cm\": 38.4}, {\"object\": \"envelope\", \"relation\": \"side_below\", \"distance_cm\": 41.0}]}", "target": "toaster oven", "split": "train"}
{"id": "scene0483_00_0", "scene_id": "scene0483_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "train"}
{"id": "scene0483_00_1", "scene_id": "scene0483_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. chair\nC. keyboard\nD. monitor", "choices": ["A. desk", "B. chair", "C. keyboard", "D. monitor"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0485_00_0", "scene_id": "scene0485_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "No", "target": "hair dryer", "split": "train"}
{"id": "scene0485_00_1", "scene_id": "scene0485_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bathroom vanity\nB. monitor\nC. toilet\nD. sink", "choices": ["A. bathroom vanity", "B. monitor", "C. toilet", "D. sink"], "answer": "D. sink", "target": "toilet paper", "split": "train"}
{"id": "scene0485_00_2", "scene_id": "scene0485_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the sink, about 36 cm away. It is also near the toilet, about 40 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0485_00_3", "scene_id": "scene0485_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom vanity\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 36.3}, {\"object\": \"toilet\", \"relation\": \"near\", \"distance_cm\": 40.0}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0486_00_0", "scene_id": "scene0486_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0486_00_1", "scene_id": "scene0486_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. food container\nB. counter\nC. kitchen cabinets\nD. paper towel roll", "choices": ["A. food container", "B. counter", "C. kitchen cabinets", "D. paper towel roll"], "answer": "C. kitchen cabinets", "target": "cup", "split": "train"}
{"id": "scene0486_00_2", "scene_id": "scene0486_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the kitchen cabinets, about 10 cm away. It is also to the side of and below the paper towel roll, about 69 cm away.", "target": "cup", "split": "train"}
{"id": "scene0486_00_3", "scene_id": "scene0486_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"paper towel roll\", \"references\": [{\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 10.3}, {\"object\": \"paper towel roll\", \"relation\": \"side_below\", \"distance_cm\": 69.3}]}", "target": "cup", "split": "train"}
{"id": "scene0489_00_0", "scene_id": "scene0489_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the fire extinguisher appear?", "choices": null, "answer": "No", "target": "fire extinguisher", "split": "train"}
{"id": "scene0489_00_1", "scene_id": "scene0489_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. books\nB. chair\nC. book\nD. bookshelf", "choices": ["A. books", "B. chair", "C. book", "D. bookshelf"], "answer": "C. book", "target": "backpack", "split": "train"}
{"id": "scene0489_00_2", "scene_id": "scene0489_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the book, about 26 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0489_00_3", "scene_id": "scene0489_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 25.7}]}", "target": "backpack", "split": "train"}
{"id": "scene0492_00_0", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper towel roll appear?", "choices": null, "answer": "Yes", "target": "paper towel roll", "split": "train"}
{"id": "scene0492_00_1", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper towel roll in this scene?\nA. sink\nB. paper towel dispenser\nC. bar\nD. trash can", "choices": ["A. sink", "B. paper towel dispenser", "C. bar", "D. trash can"], "answer": "A. sink", "target": "paper towel roll", "split": "train"}
{"id": "scene0492_00_2", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper towel roll in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper towel roll is to the side of and above the sink, about 17 cm away. It is also to the side of and above the trash can, about 43 cm away. It is further to the side of and below the paper towel dispenser, about 49 cm away.", "target": "paper towel roll", "split": "train"}
{"id": "scene0492_00_3", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"paper towel dispenser\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 16.5}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 42.6}]}", "target": "paper towel roll", "split": "train"}
{"id": "scene0492_00_4", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "scene0492_00_5", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bar\nB. paper towel roll\nC. paper towel dispenser\nD. toilet", "choices": ["A. bar", "B. paper towel roll", "C. paper towel dispenser", "D. toilet"], "answer": "A. bar", "target": "toilet paper", "split": "train"}
{"id": "scene0492_00_6", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the bar, about 7 cm away. It is also to the side of and above the toilet, about 39 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0492_00_7", "scene_id": "scene0492_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bar\", \"references\": [{\"object\": \"bar\", \"relation\": \"side_below\", \"distance_cm\": 6.9}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 39.1}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0493_00_0", "scene_id": "scene0493_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0493_00_1", "scene_id": "scene0493_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. chair\nC. monitor\nD. keyboard", "choices": ["A. desk", "B. chair", "C. monitor", "D. keyboard"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0495_00_0", "scene_id": "scene0495_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0495_00_1", "scene_id": "scene0495_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. chair\nC. desk\nD. keyboard", "choices": ["A. monitor", "B. chair", "C. desk", "D. keyboard"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "scene0495_00_2", "scene_id": "scene0495_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the chair, about 12 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0495_00_3", "scene_id": "scene0495_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 12.1}]}", "target": "backpack", "split": "train"}
{"id": "scene0497_00_0", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0497_00_1", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. table\nB. kitchen cabinets\nC. chair\nD. armchair", "choices": ["A. table", "B. kitchen cabinets", "C. chair", "D. armchair"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "scene0497_00_2", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 13 cm away. It is also to the side of and below the table, about 16 cm away. It is further to the side of and below the armchair, about 41 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0497_00_3", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 13.3}, {\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 15.9}]}", "target": "backpack", "split": "train"}
{"id": "scene0497_00_4", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0497_00_5", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. kitchen counter\nB. kitchen cabinets\nC. sink\nD. paper towel dispenser", "choices": ["A. kitchen counter", "B. kitchen cabinets", "C. sink", "D. paper towel dispenser"], "answer": "B. kitchen cabinets", "target": "soap dispenser", "split": "train"}
{"id": "scene0497_00_6", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and below the kitchen cabinets, about 16 cm away. It is also to the side of and above the kitchen counter, about 18 cm away. It is further to the side of and above the sink, about 20 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0497_00_7", "scene_id": "scene0497_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen cabinets\", \"relation\": \"side_below\", \"distance_cm\": 15.7}, {\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 18.0}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0498_00_0", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0498_00_1", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. box\nB. table\nC. boxes\nD. cup", "choices": ["A. box", "B. table", "C. boxes", "D. cup"], "answer": "B. table", "target": "bag", "split": "train"}
{"id": "scene0498_00_2", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the table, about 33 cm away. It is also to the side of and below the box, about 48 cm away. It is further to the side of and below the boxes, about 51 cm away.", "target": "bag", "split": "train"}
{"id": "scene0498_00_3", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 33.1}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 48.5}]}", "target": "bag", "split": "train"}
{"id": "scene0498_00_4", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the plate appear?", "choices": null, "answer": "Yes", "target": "plate", "split": "train"}
{"id": "scene0498_00_5", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plate in this scene?\nA. refrigerator\nB. kitchen cabinets\nC. sink\nD. bag", "choices": ["A. refrigerator", "B. kitchen cabinets", "C. sink", "D. bag"], "answer": "B. kitchen cabinets", "target": "plate", "split": "train"}
{"id": "scene0498_00_6", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plate in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plate is to the side of and above the kitchen cabinets, about 8 cm away. It is also to the side of and below the refrigerator, about 10 cm away. It is further to the side of and above the sink, about 48 cm away.", "target": "plate", "split": "train"}
{"id": "scene0498_00_7", "scene_id": "scene0498_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plate in structured JSON format.", "choices": null, "answer": "{\"target\": \"plate\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 8.5}, {\"object\": \"refrigerator\", \"relation\": \"side_below\", \"distance_cm\": 10.2}]}", "target": "plate", "split": "train"}
{"id": "scene0501_00_0", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0501_00_1", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. bottle\nB. sink\nC. bar\nD. bathroom vanity", "choices": ["A. bottle", "B. sink", "C. bar", "D. bathroom vanity"], "answer": "A. bottle", "target": "bag", "split": "train"}
{"id": "scene0501_00_2", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is next to the bottle, about 14 cm away.", "target": "bag", "split": "train"}
{"id": "scene0501_00_3", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"sink\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 13.8}]}", "target": "bag", "split": "train"}
{"id": "scene0501_00_4", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "train"}
{"id": "scene0501_00_5", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scale in this scene?\nA. toilet\nB. bag\nC. towel\nD. toilet paper", "choices": ["A. toilet", "B. bag", "C. towel", "D. toilet paper"], "answer": "A. toilet", "target": "scale", "split": "train"}
{"id": "scene0501_00_6", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is to the side of and below the toilet, about 34 cm away. It is also to the side of and below the towel, about 47 cm away. It is further to the side of and below the toilet paper, about 49 cm away.", "target": "scale", "split": "train"}
{"id": "scene0501_00_7", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 34.5}, {\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 46.9}]}", "target": "scale", "split": "train"}
{"id": "scene0501_00_8", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0501_00_9", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. scale\nB. stand\nC. toilet\nD. towel", "choices": ["A. scale", "B. stand", "C. toilet", "D. towel"], "answer": "C. toilet", "target": "toilet paper", "split": "train"}
{"id": "scene0501_00_10", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the toilet, about 34 cm away. It is also to the side of and above the scale, about 49 cm away. It is further to the side of and above the stand, about 65 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0501_00_11", "scene_id": "scene0501_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"stand\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 33.6}, {\"object\": \"scale\", \"relation\": \"side_above\", \"distance_cm\": 48.6}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0504_00_0", "scene_id": "scene0504_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0504_00_1", "scene_id": "scene0504_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. mirror\nB. monitor\nC. chair\nD. desk", "choices": ["A. mirror", "B. monitor", "C. chair", "D. desk"], "answer": "A. mirror", "target": "bag", "split": "train"}
{"id": "scene0504_00_2", "scene_id": "scene0504_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and below the mirror, about 28 cm away.", "target": "bag", "split": "train"}
{"id": "scene0504_00_3", "scene_id": "scene0504_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"mirror\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 28.1}]}", "target": "bag", "split": "train"}
{"id": "scene0510_00_0", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "scene0510_00_1", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. hand dryer\nB. shower control valve\nC. bar\nD. jacket", "choices": ["A. hand dryer", "B. shower control valve", "C. bar", "D. jacket"], "answer": "B. shower control valve", "target": "bottle", "split": "train"}
{"id": "scene0510_00_2", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the shower control valve, about 74 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0510_00_3", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shower walls\", \"references\": [{\"object\": \"shower control valve\", \"relation\": \"side_below\", \"distance_cm\": 73.6}]}", "target": "bottle", "split": "train"}
{"id": "scene0510_00_4", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "scene0510_00_5", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. bottle\nB. toilet\nC. bar\nD. trash bin", "choices": ["A. bottle", "B. toilet", "C. bar", "D. trash bin"], "answer": "D. trash bin", "target": "toilet paper", "split": "train"}
{"id": "scene0510_00_6", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is above the trash bin, about 10 cm away. It is also to the side of and below the bar, about 14 cm away. It is further to the side of and above the toilet, about 35 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0510_00_7", "scene_id": "scene0510_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"trash bin\", \"references\": [{\"object\": \"trash bin\", \"relation\": \"above\", \"distance_cm\": 9.7}, {\"object\": \"bar\", \"relation\": \"side_below\", \"distance_cm\": 13.8}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0515_00_0", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the telephone appear?", "choices": null, "answer": "No", "target": "telephone", "split": "train"}
{"id": "scene0515_00_1", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the guitar in this scene?\nA. backpack\nB. bathroom vanity\nC. keyboard piano\nD. radiator", "choices": ["A. backpack", "B. bathroom vanity", "C. keyboard piano", "D. radiator"], "answer": "C. keyboard piano", "target": "guitar", "split": "train"}
{"id": "scene0515_00_2", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the guitar in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The guitar is to the side of and below the keyboard piano, about 20 cm away. It is also to the side of and above the radiator, about 24 cm away.", "target": "guitar", "split": "train"}
{"id": "scene0515_00_3", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the guitar in structured JSON format.", "choices": null, "answer": "{\"target\": \"guitar\", \"support_surface\": \"radiator\", \"references\": [{\"object\": \"keyboard piano\", \"relation\": \"side_below\", \"distance_cm\": 19.6}, {\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 24.1}]}", "target": "guitar", "split": "train"}
{"id": "scene0515_00_4", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0515_00_5", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. chair\nB. backpack\nC. bathroom vanity\nD. bottle", "choices": ["A. chair", "B. backpack", "C. bathroom vanity", "D. bottle"], "answer": "A. chair", "target": "laptop", "split": "train"}
{"id": "scene0515_00_6", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is on the desk. It is also to the side of and above the chair, about 27 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0515_00_7", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 27.4}]}", "target": "laptop", "split": "train"}
{"id": "scene0515_00_8", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "No", "target": "hair dryer", "split": "train"}
{"id": "scene0515_00_9", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scale in this scene?\nA. bathroom vanity\nB. mirror\nC. backpack\nD. sink", "choices": ["A. bathroom vanity", "B. mirror", "C. backpack", "D. sink"], "answer": "A. bathroom vanity", "target": "scale", "split": "train"}
{"id": "scene0515_00_10", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is to the side of and below the bathroom vanity, about 50 cm away. It is also to the side of and below the sink, about 65 cm away. It is further to the side of and below the mirror, about 82 cm away.", "target": "scale", "split": "train"}
{"id": "scene0515_00_11", "scene_id": "scene0515_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"bathroom vanity\", \"references\": [{\"object\": \"bathroom vanity\", \"relation\": \"side_below\", \"distance_cm\": 50.2}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 65.3}]}", "target": "scale", "split": "train"}
{"id": "scene0516_00_0", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bag appear?", "choices": null, "answer": "No", "target": "bag", "split": "train"}
{"id": "scene0516_00_1", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. bin\nB. mirror\nC. picture\nD. toilet", "choices": ["A. bin", "B. mirror", "C. picture", "D. toilet"], "answer": "B. mirror", "target": "bottle", "split": "train"}
{"id": "scene0516_00_2", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the mirror, about 18 cm away. It is also to the side of and above the bin, about 27 cm away. It is further to the side of and above the toilet, about 39 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0516_00_3", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bin\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 17.8}, {\"object\": \"bin\", \"relation\": \"side_above\", \"distance_cm\": 26.7}]}", "target": "bottle", "split": "train"}
{"id": "scene0516_00_4", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "train"}
{"id": "scene0516_00_5", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toilet paper in this scene?\nA. trash can\nB. bottle\nC. bin\nD. toilet", "choices": ["A. trash can", "B. bottle", "C. bin", "D. toilet"], "answer": "A. trash can", "target": "toilet paper", "split": "train"}
{"id": "scene0516_00_6", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and above the trash can, about 12 cm away. It is also to the side of and above the toilet, about 34 cm away. It is further to the side of and above the bin, about 55 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0516_00_7", "scene_id": "scene0516_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bin\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 11.9}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 33.9}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0517_00_0", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dumbbell appear?", "choices": null, "answer": "No", "target": "dumbbell", "split": "train"}
{"id": "scene0517_00_1", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bag in this scene?\nA. bottle\nB. towel\nC. backpack\nD. sink", "choices": ["A. bottle", "B. towel", "C. backpack", "D. sink"], "answer": "D. sink", "target": "bag", "split": "train"}
{"id": "scene0517_00_2", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is to the side of and above the sink, about 12 cm away. It is also to the side of and above the backpack, about 20 cm away. It is further to the side of and below the bottle, about 22 cm away.", "target": "bag", "split": "train"}
{"id": "scene0517_00_3", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"backpack\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 11.7}, {\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 20.3}]}", "target": "bag", "split": "train"}
{"id": "scene0517_00_4", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "scene0517_00_5", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the sink, about 10 cm away. It is also to the side of and below the mirror, about 11 cm away. It is further to the side of and above the bag, about 22 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0517_00_6", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 10.5}, {\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 10.6}]}", "target": "bottle", "split": "train"}
{"id": "scene0517_00_7", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "Yes", "target": "bowl", "split": "train"}
{"id": "scene0517_00_8", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bowl in this scene?\nA. laptop\nB. chair\nC. picture\nD. mug", "choices": ["A. laptop", "B. chair", "C. picture", "D. mug"], "answer": "A. laptop", "target": "bowl", "split": "train"}
{"id": "scene0517_00_9", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bowl in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bowl is on the desk. It is also near the laptop, about 20 cm away. It is further to the side of and above the chair, about 29 cm away.", "target": "bowl", "split": "train"}
{"id": "scene0517_00_10", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bowl in structured JSON format.", "choices": null, "answer": "{\"target\": \"bowl\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 19.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 28.5}]}", "target": "bowl", "split": "train"}
{"id": "scene0517_00_11", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0517_00_12", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water pitcher in this scene?\nA. flip flops\nB. shoes\nC. tissue box\nD. chair", "choices": ["A. flip flops", "B. shoes", "C. tissue box", "D. chair"], "answer": "A. flip flops", "target": "water pitcher", "split": "train"}
{"id": "scene0517_00_13", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water pitcher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water pitcher is on the coffee table. It is also to the side of and above the flip flops, about 47 cm away. It is further near the tissue box, about 51 cm away.", "target": "water pitcher", "split": "train"}
{"id": "scene0517_00_14", "scene_id": "scene0517_00", "source_dataset": "scannet200", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water pitcher in structured JSON format.", "choices": null, "answer": "{\"target\": \"water pitcher\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"flip flops\", \"relation\": \"side_above\", \"distance_cm\": 46.9}, {\"object\": \"tissue box\", \"relation\": \"near\", \"distance_cm\": 50.5}]}", "target": "water pitcher", "split": "train"}
{"id": "scene0519_00_0", "scene_id": "scene0519_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "No", "target": "toaster", "split": "train"}
{"id": "scene0519_00_1", "scene_id": "scene0519_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. trash bin\nB. bathroom vanity\nC. mirror\nD. paper towel dispenser", "choices": ["A. trash bin", "B. bathroom vanity", "C. mirror", "D. paper towel dispenser"], "answer": "D. paper towel dispenser", "target": "soap dispenser", "split": "train"}
{"id": "scene0519_00_2", "scene_id": "scene0519_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and below the paper towel dispenser, about 9 cm away. It is also to the side of and above the mirror, about 15 cm away. It is further to the side of and above the bathroom vanity, about 19 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0519_00_3", "scene_id": "scene0519_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"paper towel dispenser\", \"references\": [{\"object\": \"paper towel dispenser\", \"relation\": \"side_below\", \"distance_cm\": 8.8}, {\"object\": \"mirror\", \"relation\": \"side_above\", \"distance_cm\": 14.7}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0520_00_0", "scene_id": "scene0520_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper towel roll appear?", "choices": null, "answer": "No", "target": "paper towel roll", "split": "train"}
{"id": "scene0520_00_1", "scene_id": "scene0520_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. monitor\nB. armchair\nC. table\nD. picture", "choices": ["A. monitor", "B. armchair", "C. table", "D. picture"], "answer": "B. armchair", "target": "book", "split": "train"}
{"id": "scene0520_00_2", "scene_id": "scene0520_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is on the table. It is also near the armchair, about 25 cm away. It is further to the side of and below the picture, about 54 cm away.", "target": "book", "split": "train"}
{"id": "scene0520_00_3", "scene_id": "scene0520_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"table\", \"references\": [{\"object\": \"armchair\", \"relation\": \"near\", \"distance_cm\": 24.5}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 54.0}]}", "target": "book", "split": "train"}
{"id": "scene0521_00_0", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0521_00_1", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. books\nB. box\nC. hat\nD. bookshelf", "choices": ["A. books", "B. box", "C. hat", "D. bookshelf"], "answer": "A. books", "target": "book", "split": "train"}
{"id": "scene0521_00_2", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is under the books, about 3 cm away. It is also next to the hat, about 7 cm away. It is further to the side of and above the box, about 15 cm away.", "target": "book", "split": "train"}
{"id": "scene0521_00_3", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"books\", \"relation\": \"under\", \"distance_cm\": 3.2}, {\"object\": \"hat\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "book", "split": "train"}
{"id": "scene0521_00_4", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "train"}
{"id": "scene0521_00_5", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hat in this scene?\nA. book\nB. box\nC. books\nD. bookshelf", "choices": ["A. book", "B. box", "C. books", "D. bookshelf"], "answer": "A. book", "target": "hat", "split": "train"}
{"id": "scene0521_00_6", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hat in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hat is next to the book, about 7 cm away. It is also to the side of and below the books, about 19 cm away. It is further to the side of and above the box, about 38 cm away.", "target": "hat", "split": "train"}
{"id": "scene0521_00_7", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hat in structured JSON format.", "choices": null, "answer": "{\"target\": \"hat\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 7.1}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 19.4}]}", "target": "hat", "split": "train"}
{"id": "scene0521_00_8", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "scene0521_00_9", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. towel\nB. toilet\nC. picture\nD. toilet paper", "choices": ["A. towel", "B. toilet", "C. picture", "D. toilet paper"], "answer": "C. picture", "target": "tissue box", "split": "train"}
{"id": "scene0521_00_10", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and below the picture, about 10 cm away. It is also to the side of and above the toilet, about 19 cm away. It is further to the side of and above the towel, about 22 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0521_00_11", "scene_id": "scene0521_00", "source_dataset": "scannet200", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 9.7}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 19.3}]}", "target": "tissue box", "split": "train"}
{"id": "scene0523_00_0", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "Yes", "target": "hair dryer", "split": "train"}
{"id": "scene0523_00_1", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hair dryer in this scene?\nA. towel\nB. trash can\nC. sink\nD. toilet paper", "choices": ["A. towel", "B. trash can", "C. sink", "D. toilet paper"], "answer": "D. toilet paper", "target": "hair dryer", "split": "train"}
{"id": "scene0523_00_2", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hair dryer in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hair dryer is above the toilet paper, about 8 cm away. It is also to the side of and above the towel, about 9 cm away. It is further to the side of and above the trash can, about 36 cm away.", "target": "hair dryer", "split": "train"}
{"id": "scene0523_00_3", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hair dryer in structured JSON format.", "choices": null, "answer": "{\"target\": \"hair dryer\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"above\", \"distance_cm\": 7.8}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 9.4}]}", "target": "hair dryer", "split": "train"}
{"id": "scene0523_00_4", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the alarm clock appear?", "choices": null, "answer": "No", "target": "alarm clock", "split": "train"}
{"id": "scene0523_00_5", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. mirror\nB. towel\nC. sink\nD. bathroom vanity", "choices": ["A. mirror", "B. towel", "C. sink", "D. bathroom vanity"], "answer": "D. bathroom vanity", "target": "tissue box", "split": "train"}
{"id": "scene0523_00_6", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and above the bathroom vanity, about 5 cm away. It is also to the side of and below the mirror, about 6 cm away. It is further to the side of and above the sink, about 17 cm away.", "target": "tissue box", "split": "train"}
{"id": "scene0523_00_7", "scene_id": "scene0523_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"bathroom vanity\", \"relation\": \"side_above\", \"distance_cm\": 4.7}, {\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 6.5}]}", "target": "tissue box", "split": "train"}
{"id": "scene0525_00_0", "scene_id": "scene0525_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0525_00_1", "scene_id": "scene0525_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tray in this scene?\nA. office chair\nB. radiator\nC. book\nD. plant", "choices": ["A. office chair", "B. radiator", "C. book", "D. plant"], "answer": "B. radiator", "target": "tray", "split": "train"}
{"id": "scene0525_00_2", "scene_id": "scene0525_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tray in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tray is on the table. It is also to the side of and above the radiator, about 21 cm away. It is further to the side of and above the office chair, about 33 cm away.", "target": "tray", "split": "train"}
{"id": "scene0525_00_3", "scene_id": "scene0525_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tray in structured JSON format.", "choices": null, "answer": "{\"target\": \"tray\", \"support_surface\": \"table\", \"references\": [{\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 20.9}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 32.6}]}", "target": "tray", "split": "train"}
{"id": "scene0530_00_0", "scene_id": "scene0530_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the guitar appear?", "choices": null, "answer": "No", "target": "guitar", "split": "train"}
{"id": "scene0530_00_1", "scene_id": "scene0530_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the fire extinguisher appear?", "choices": null, "answer": "Yes", "target": "fire extinguisher", "split": "train"}
{"id": "scene0530_00_2", "scene_id": "scene0530_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the fire extinguisher in this scene?\nA. monitor\nB. chair\nC. basket\nD. backpack", "choices": ["A. monitor", "B. chair", "C. basket", "D. backpack"], "answer": "C. basket", "target": "fire extinguisher", "split": "train"}
{"id": "scene0530_00_3", "scene_id": "scene0530_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the fire extinguisher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The fire extinguisher is to the side of and below the basket, about 44 cm away.", "target": "fire extinguisher", "split": "train"}
{"id": "scene0530_00_4", "scene_id": "scene0530_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the fire extinguisher in structured JSON format.", "choices": null, "answer": "{\"target\": \"fire extinguisher\", \"support_surface\": \"basket\", \"references\": [{\"object\": \"basket\", \"relation\": \"side_below\", \"distance_cm\": 43.8}]}", "target": "fire extinguisher", "split": "train"}
{"id": "scene0534_00_0", "scene_id": "scene0534_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "scene0534_00_1", "scene_id": "scene0534_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cabinets\nB. counter\nC. water cooler\nD. bulletin board", "choices": ["A. cabinets", "B. counter", "C. water cooler", "D. bulletin board"], "answer": "C. water cooler", "target": "backpack", "split": "train"}
{"id": "scene0534_00_2", "scene_id": "scene0534_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the shelf. It is also to the side of and above the water cooler, about 46 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0534_00_3", "scene_id": "scene0534_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"water cooler\", \"relation\": \"side_above\", \"distance_cm\": 45.8}]}", "target": "backpack", "split": "train"}
{"id": "scene0534_00_4", "scene_id": "scene0534_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "Yes", "target": "paper cutter", "split": "train"}
{"id": "scene0534_00_5", "scene_id": "scene0534_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper cutter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper cutter is to the side of and below the windowsill, about 6 cm away. It is also to the side of and above the cabinets, about 7 cm away.", "target": "paper cutter", "split": "train"}
{"id": "scene0534_00_6", "scene_id": "scene0534_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper cutter in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper cutter\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"cabinets\", \"relation\": \"side_above\", \"distance_cm\": 7.0}]}", "target": "paper cutter", "split": "train"}
{"id": "scene0539_00_0", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the clock appear?", "choices": null, "answer": "No", "target": "clock", "split": "train"}
{"id": "scene0539_00_1", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mirror\nB. plant\nC. sink\nD. hair dryer", "choices": ["A. mirror", "B. plant", "C. sink", "D. hair dryer"], "answer": "A. mirror", "target": "bottle", "split": "train"}
{"id": "scene0539_00_2", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the mirror, about 28 cm away. It is also to the side of and above the hair dryer, about 35 cm away. It is further to the side of and above the sink, about 52 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0539_00_3", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 28.4}, {\"object\": \"hair dryer\", \"relation\": \"side_above\", \"distance_cm\": 35.4}]}", "target": "bottle", "split": "train"}
{"id": "scene0539_00_4", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "Yes", "target": "hair dryer", "split": "train"}
{"id": "scene0539_00_5", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hair dryer in this scene?\nA. bathroom vanity\nB. plant\nC. bottle\nD. sink", "choices": ["A. bathroom vanity", "B. plant", "C. bottle", "D. sink"], "answer": "A. bathroom vanity", "target": "hair dryer", "split": "train"}
{"id": "scene0539_00_6", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hair dryer in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hair dryer is to the side of and above the bathroom vanity, about 29 cm away. It is also to the side of and below the bottle, about 35 cm away. It is further near the sink, about 40 cm away.", "target": "hair dryer", "split": "train"}
{"id": "scene0539_00_7", "scene_id": "scene0539_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hair dryer in structured JSON format.", "choices": null, "answer": "{\"target\": \"hair dryer\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bathroom vanity\", \"relation\": \"side_above\", \"distance_cm\": 29.1}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 35.4}]}", "target": "hair dryer", "split": "train"}
{"id": "scene0540_00_0", "scene_id": "scene0540_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0540_00_1", "scene_id": "scene0540_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. spray bottle\nB. stool\nC. monitor\nD. table", "choices": ["A. spray bottle", "B. stool", "C. monitor", "D. table"], "answer": "B. stool", "target": "telephone", "split": "train"}
{"id": "scene0540_00_2", "scene_id": "scene0540_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the stool, about 16 cm away. It is also to the side of and above the table, about 24 cm away. It is further to the side of and above the spray bottle, about 82 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0540_00_3", "scene_id": "scene0540_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"stool\", \"relation\": \"side_above\", \"distance_cm\": 16.3}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 24.3}]}", "target": "telephone", "split": "train"}
{"id": "scene0542_00_0", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "Yes", "target": "bowl", "split": "train"}
{"id": "scene0542_00_1", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bowl in this scene?\nA. clock\nB. sink\nC. kitchen cabinets\nD. kitchen counter", "choices": ["A. clock", "B. sink", "C. kitchen cabinets", "D. kitchen counter"], "answer": "D. kitchen counter", "target": "bowl", "split": "train"}
{"id": "scene0542_00_2", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bowl in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bowl is to the side of and above the kitchen counter, about 11 cm away. It is also to the side of and above the sink, about 13 cm away. It is further to the side of and above the kitchen cabinets, about 37 cm away.", "target": "bowl", "split": "train"}
{"id": "scene0542_00_3", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bowl in structured JSON format.", "choices": null, "answer": "{\"target\": \"bowl\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 10.9}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 13.4}]}", "target": "bowl", "split": "train"}
{"id": "scene0542_00_4", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0542_00_5", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the clock in this scene?\nA. bowl\nB. range hood\nC. shelf\nD. kitchen cabinets", "choices": ["A. bowl", "B. range hood", "C. shelf", "D. kitchen cabinets"], "answer": "C. shelf", "target": "clock", "split": "train"}
{"id": "scene0542_00_6", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the clock in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The clock is to the side of and above the shelf, about 7 cm away. It is also near the range hood, about 51 cm away.", "target": "clock", "split": "train"}
{"id": "scene0542_00_7", "scene_id": "scene0542_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"clock\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 7.1}, {\"object\": \"range hood\", \"relation\": \"near\", \"distance_cm\": 51.1}]}", "target": "clock", "split": "train"}
{"id": "scene0548_00_0", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bowl appear?", "choices": null, "answer": "No", "target": "bowl", "split": "train"}
{"id": "scene0548_00_1", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is on the bookshelf. It is also to the side of and above the book, about 30 cm away.", "target": "bag", "split": "train"}
{"id": "scene0548_00_2", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 29.8}]}", "target": "bag", "split": "train"}
{"id": "scene0548_00_3", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "train"}
{"id": "scene0548_00_4", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. bookshelf\nB. books\nC. bag\nD. cabinet", "choices": ["A. bookshelf", "B. books", "C. bag", "D. cabinet"], "answer": "B. books", "target": "book", "split": "train"}
{"id": "scene0548_00_5", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the book in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The book is to the side of and below the bag, about 30 cm away.", "target": "book", "split": "train"}
{"id": "scene0548_00_6", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 29.8}]}", "target": "book", "split": "train"}
{"id": "scene0548_00_7", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "scene0548_00_8", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. chair\nB. cabinet\nC. bag\nD. lamp", "choices": ["A. chair", "B. cabinet", "C. bag", "D. lamp"], "answer": "B. cabinet", "target": "cup", "split": "train"}
{"id": "scene0548_00_9", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also to the side of and above the cabinet, about 14 cm away. It is further to the side of and above the chair, about 29 cm away.", "target": "cup", "split": "train"}
{"id": "scene0548_00_10", "scene_id": "scene0548_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 13.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 28.6}]}", "target": "cup", "split": "train"}
{"id": "scene0552_00_0", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dumbbell appear?", "choices": null, "answer": "No", "target": "dumbbell", "split": "train"}
{"id": "scene0552_00_1", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper cutter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper cutter is on the table. It is also to the side of and above the trash can, about 36 cm away. It is further to the side of and above the bucket, about 37 cm away.", "target": "paper cutter", "split": "train"}
{"id": "scene0552_00_2", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper cutter in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper cutter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 36.5}, {\"object\": \"bucket\", \"relation\": \"side_above\", \"distance_cm\": 37.4}]}", "target": "paper cutter", "split": "train"}
{"id": "scene0552_00_3", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "No", "target": "book", "split": "train"}
{"id": "scene0552_00_4", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. toaster oven\nB. paper towel dispenser\nC. sink\nD. coffee maker", "choices": ["A. toaster oven", "B. paper towel dispenser", "C. sink", "D. coffee maker"], "answer": "C. sink", "target": "soap dispenser", "split": "train"}
{"id": "scene0552_00_5", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the sink, about 9 cm away. It is also to the side of and above the coffee maker, about 17 cm away. It is further to the side of and below the paper towel dispenser, about 23 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0552_00_6", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 9.3}, {\"object\": \"coffee maker\", \"relation\": \"side_above\", \"distance_cm\": 16.7}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0552_00_7", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster oven appear?", "choices": null, "answer": "Yes", "target": "toaster oven", "split": "train"}
{"id": "scene0552_00_8", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toaster oven in this scene?\nA. kitchen cabinets\nB. coffee maker\nC. soap dispenser\nD. bucket", "choices": ["A. kitchen cabinets", "B. coffee maker", "C. soap dispenser", "D. bucket"], "answer": "B. coffee maker", "target": "toaster oven", "split": "train"}
{"id": "scene0552_00_9", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toaster oven in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster oven is to the side of and below the coffee maker, about 11 cm away. It is also to the side of and above the kitchen cabinets, about 26 cm away. It is further to the side of and below the soap dispenser, about 50 cm away.", "target": "toaster oven", "split": "train"}
{"id": "scene0552_00_10", "scene_id": "scene0552_00", "source_dataset": "scannet200", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toaster oven in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster oven\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"coffee maker\", \"relation\": \"side_below\", \"distance_cm\": 11.2}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 25.8}]}", "target": "toaster oven", "split": "train"}
{"id": "scene0553_00_0", "scene_id": "scene0553_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scale appear?", "choices": null, "answer": "No", "target": "scale", "split": "train"}
{"id": "scene0553_00_1", "scene_id": "scene0553_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "Yes", "target": "soap dispenser", "split": "train"}
{"id": "scene0553_00_2", "scene_id": "scene0553_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the sink, about 23 cm away. It is also to the side of and below the mirror, about 24 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0553_00_3", "scene_id": "scene0553_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"sink\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 22.9}, {\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 23.8}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0556_00_0", "scene_id": "scene0556_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "scene0556_00_1", "scene_id": "scene0556_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the candle appear?", "choices": null, "answer": "No", "target": "candle", "split": "train"}
{"id": "scene0556_00_2", "scene_id": "scene0556_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"dresser\", \"references\": []}", "target": "telephone", "split": "train"}
{"id": "scene0557_00_0", "scene_id": "scene0557_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "scene0557_00_1", "scene_id": "scene0557_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"nightstand\", \"references\": []}", "target": "book", "split": "train"}
{"id": "scene0558_00_0", "scene_id": "scene0558_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toilet paper appear?", "choices": null, "answer": "No", "target": "toilet paper", "split": "train"}
{"id": "scene0558_00_1", "scene_id": "scene0558_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0558_00_2", "scene_id": "scene0558_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the book in this scene?\nA. books\nB. bookshelf\nC. chair\nD. backpack", "choices": ["A. books", "B. bookshelf", "C. chair", "D. backpack"], "answer": "A. books", "target": "book", "split": "train"}
{"id": "scene0558_00_3", "scene_id": "scene0558_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the book in structured JSON format.", "choices": null, "answer": "{\"target\": \"book\", \"support_surface\": \"bookshelf\", \"references\": []}", "target": "book", "split": "train"}
{"id": "scene0562_00_0", "scene_id": "scene0562_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dumbbell appear?", "choices": null, "answer": "No", "target": "dumbbell", "split": "train"}
{"id": "scene0562_00_1", "scene_id": "scene0562_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. table\nB. desk\nC. chair\nD. monitor", "choices": ["A. table", "B. desk", "C. chair", "D. monitor"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "scene0562_00_2", "scene_id": "scene0562_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and above the chair, about 60 cm away.", "target": "backpack", "split": "train"}
{"id": "scene0562_00_3", "scene_id": "scene0562_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 59.6}]}", "target": "backpack", "split": "train"}
{"id": "scene0565_00_0", "scene_id": "scene0565_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "train"}
{"id": "scene0565_00_1", "scene_id": "scene0565_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tray appear?", "choices": null, "answer": "No", "target": "tray", "split": "train"}
{"id": "scene0565_00_2", "scene_id": "scene0565_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. computer tower\nB. printer\nC. monitor\nD. chair", "choices": ["A. computer tower", "B. printer", "C. monitor", "D. chair"], "answer": "B. printer", "target": "keyboard", "split": "train"}
{"id": "scene0565_00_3", "scene_id": "scene0565_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the printer, about 4 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0565_00_4", "scene_id": "scene0565_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"printer\", \"relation\": \"under\", \"distance_cm\": 4.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.5}]}", "target": "keyboard", "split": "train"}
{"id": "scene0569_00_0", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "train"}
{"id": "scene0569_00_1", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. keyboard\nB. monitor\nC. chair\nD. laptop", "choices": ["A. keyboard", "B. monitor", "C. chair", "D. laptop"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "scene0569_00_2", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop appear?", "choices": null, "answer": "Yes", "target": "laptop", "split": "train"}
{"id": "scene0569_00_3", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop in this scene?\nA. backpack\nB. chair\nC. monitor\nD. mouse", "choices": ["A. backpack", "B. chair", "C. monitor", "D. mouse"], "answer": "C. monitor", "target": "laptop", "split": "train"}
{"id": "scene0569_00_4", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop is to the side of and below the monitor, about 6 cm away. It is also near the mouse, about 45 cm away.", "target": "laptop", "split": "train"}
{"id": "scene0569_00_5", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.8}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 45.4}]}", "target": "laptop", "split": "train"}
{"id": "scene0569_00_6", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "train"}
{"id": "scene0569_00_7", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. laptop\nB. monitor\nC. chair\nD. keyboard", "choices": ["A. laptop", "B. monitor", "C. chair", "D. keyboard"], "answer": "B. monitor", "target": "mouse", "split": "train"}
{"id": "scene0569_00_8", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and below the monitor, about 13 cm away. It is also to the side of and above the chair, about 19 cm away. It is further near the laptop, about 45 cm away.", "target": "mouse", "split": "train"}
{"id": "scene0569_00_9", "scene_id": "scene0569_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 18.8}]}", "target": "mouse", "split": "train"}
{"id": "scene0570_00_0", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0570_00_1", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the plunger in this scene?\nA. sink\nB. soap bottle\nC. toilet\nD. bathroom counter", "choices": ["A. sink", "B. soap bottle", "C. toilet", "D. bathroom counter"], "answer": "C. toilet", "target": "plunger", "split": "train"}
{"id": "scene0570_00_2", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the plunger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plunger is to the side of and below the toilet, about 6 cm away. It is also to the side of and below the bathroom counter, about 36 cm away. It is further to the side of and below the soap bottle, about 49 cm away.", "target": "plunger", "split": "train"}
{"id": "scene0570_00_3", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the plunger in structured JSON format.", "choices": null, "answer": "{\"target\": \"plunger\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 6.0}, {\"object\": \"bathroom counter\", \"relation\": \"side_below\", \"distance_cm\": 35.6}]}", "target": "plunger", "split": "train"}
{"id": "scene0570_00_4", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "Yes", "target": "soap dish", "split": "train"}
{"id": "scene0570_00_5", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. shower control valve\nB. plunger\nC. bathtub\nD. bathroom counter", "choices": ["A. shower control valve", "B. plunger", "C. bathtub", "D. bathroom counter"], "answer": "C. bathtub", "target": "soap dish", "split": "train"}
{"id": "scene0570_00_6", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the bathtub, about 15 cm away. It is also to the side of and below the shower control valve, about 67 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0570_00_7", "scene_id": "scene0570_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"shower walls\", \"references\": [{\"object\": \"bathtub\", \"relation\": \"side_above\", \"distance_cm\": 14.8}, {\"object\": \"shower control valve\", \"relation\": \"side_below\", \"distance_cm\": 66.8}]}", "target": "soap dish", "split": "train"}
{"id": "scene0574_00_0", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the coffee kettle appear?", "choices": null, "answer": "No", "target": "coffee kettle", "split": "train"}
{"id": "scene0574_00_1", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hair dryer in this scene?\nA. mirror\nB. bathroom vanity\nC. towel\nD. sink", "choices": ["A. mirror", "B. bathroom vanity", "C. towel", "D. sink"], "answer": "B. bathroom vanity", "target": "hair dryer", "split": "train"}
{"id": "scene0574_00_2", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hair dryer in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hair dryer is to the side of and above the bathroom vanity. It is also to the side of and above the towel, about 7 cm away. It is further to the side of and above the mirror, about 17 cm away.", "target": "hair dryer", "split": "train"}
{"id": "scene0574_00_3", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hair dryer in structured JSON format.", "choices": null, "answer": "{\"target\": \"hair dryer\", \"support_surface\": \"towel\", \"references\": [{\"object\": \"bathroom vanity\", \"relation\": \"side_above\", \"distance_cm\": 2.2}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 6.9}]}", "target": "hair dryer", "split": "train"}
{"id": "scene0574_00_4", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "Yes", "target": "soap dish", "split": "train"}
{"id": "scene0574_00_5", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. sink\nB. towel\nC. mirror\nD. trash can", "choices": ["A. sink", "B. towel", "C. mirror", "D. trash can"], "answer": "A. sink", "target": "soap dish", "split": "train"}
{"id": "scene0574_00_6", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the sink, about 6 cm away. It is also to the side of and below the towel, about 11 cm away. It is further to the side of and below the mirror, about 13 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0574_00_7", "scene_id": "scene0574_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"sink\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 6.1}, {\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 10.8}]}", "target": "soap dish", "split": "train"}
{"id": "scene0578_00_0", "scene_id": "scene0578_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the book appear?", "choices": null, "answer": "No", "target": "book", "split": "train"}
{"id": "scene0578_00_1", "scene_id": "scene0578_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. table\nC. monitor\nD. chair", "choices": ["A. desk", "B. table", "C. monitor", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "train"}
{"id": "scene0583_00_0", "scene_id": "scene0583_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "scene0583_00_1", "scene_id": "scene0583_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mirror\nB. toilet\nC. toilet paper\nD. trash can", "choices": ["A. mirror", "B. toilet", "C. toilet paper", "D. trash can"], "answer": "C. toilet paper", "target": "bottle", "split": "train"}
{"id": "scene0583_00_2", "scene_id": "scene0583_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the toilet paper, about 20 cm away. It is also to the side of and below the mirror, about 27 cm away. It is further to the side of and above the trash can, about 39 cm away.", "target": "bottle", "split": "train"}
{"id": "scene0583_00_3", "scene_id": "scene0583_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"side_above\", \"distance_cm\": 19.6}, {\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 27.2}]}", "target": "bottle", "split": "train"}
{"id": "scene0583_00_4", "scene_id": "scene0583_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hair dryer appear?", "choices": null, "answer": "No", "target": "hair dryer", "split": "train"}
{"id": "scene0583_00_5", "scene_id": "scene0583_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toilet paper in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toilet paper is to the side of and below the bottle, about 20 cm away. It is also to the side of and above the trash can, about 20 cm away. It is further near the sink, about 22 cm away.", "target": "toilet paper", "split": "train"}
{"id": "scene0583_00_6", "scene_id": "scene0583_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom vanity\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 19.6}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 19.6}]}", "target": "toilet paper", "split": "train"}
{"id": "scene0584_00_0", "scene_id": "scene0584_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "train"}
{"id": "scene0584_00_1", "scene_id": "scene0584_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. desk\nB. chair\nC. backpack\nD. monitor", "choices": ["A. desk", "B. chair", "C. backpack", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "scene0584_00_2", "scene_id": "scene0584_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the monitor, about 8 cm away. It is also to the side of and above the chair, about 23 cm away.", "target": "keyboard", "split": "train"}
{"id": "scene0584_00_3", "scene_id": "scene0584_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 22.8}]}", "target": "keyboard", "split": "train"}
{"id": "scene0585_00_0", "scene_id": "scene0585_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "train"}
{"id": "scene0585_00_1", "scene_id": "scene0585_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the paper cutter in this scene?\nA. trash can\nB. cart\nC. monitor\nD. shelf", "choices": ["A. trash can", "B. cart", "C. monitor", "D. shelf"], "answer": "A. trash can", "target": "paper cutter", "split": "train"}
{"id": "scene0585_00_2", "scene_id": "scene0585_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the paper cutter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The paper cutter is on the shelf. It is also to the side of and above the trash can, about 20 cm away. It is further to the side of and above the cart, about 33 cm away.", "target": "paper cutter", "split": "train"}
{"id": "scene0585_00_3", "scene_id": "scene0585_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the paper cutter in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper cutter\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 19.7}, {\"object\": \"cart\", \"relation\": \"side_above\", \"distance_cm\": 32.6}]}", "target": "paper cutter", "split": "train"}
{"id": "scene0586_00_0", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cd case appear?", "choices": null, "answer": "No", "target": "cd case", "split": "train"}
{"id": "scene0586_00_1", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. soap dispenser\nB. coffee maker\nC. kitchen counter\nD. kitchen cabinets", "choices": ["A. soap dispenser", "B. coffee maker", "C. kitchen counter", "D. kitchen cabinets"], "answer": "A. soap dispenser", "target": "cup", "split": "train"}
{"id": "scene0586_00_2", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the soap dispenser, about 25 cm away. It is also to the side of and above the kitchen cabinets, about 48 cm away. It is further to the side of and below the coffee maker, about 48 cm away.", "target": "cup", "split": "train"}
{"id": "scene0586_00_3", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"soap dispenser\", \"relation\": \"side_below\", \"distance_cm\": 24.5}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 47.5}]}", "target": "cup", "split": "train"}
{"id": "scene0586_00_4", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0586_00_5", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dispenser in this scene?\nA. kitchen counter\nB. sink\nC. cup\nD. coffee maker", "choices": ["A. kitchen counter", "B. sink", "C. cup", "D. coffee maker"], "answer": "A. kitchen counter", "target": "soap dispenser", "split": "train"}
{"id": "scene0586_00_6", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dispenser is to the side of and above the kitchen counter, about 5 cm away. It is also to the side of and above the sink, about 18 cm away. It is further to the side of and above the coffee maker, about 24 cm away.", "target": "soap dispenser", "split": "train"}
{"id": "scene0586_00_7", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 4.6}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 17.5}]}", "target": "soap dispenser", "split": "train"}
{"id": "scene0586_00_8", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toaster appear?", "choices": null, "answer": "Yes", "target": "toaster", "split": "train"}
{"id": "scene0586_00_9", "scene_id": "scene0586_00", "source_dataset": "scannet200", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toaster in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster\", \"support_surface\": \"table\", \"references\": []}", "target": "toaster", "split": "train"}
{"id": "scene0702_00_0", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap dish appear?", "choices": null, "answer": "Yes", "target": "soap dish", "split": "train"}
{"id": "scene0702_00_1", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap dish in this scene?\nA. faucet\nB. shower walls\nC. bar\nD. bathtub", "choices": ["A. faucet", "B. shower walls", "C. bar", "D. bathtub"], "answer": "D. bathtub", "target": "soap dish", "split": "train"}
{"id": "scene0702_00_2", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap dish in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap dish is to the side of and above the bathtub, about 6 cm away. It is also to the side of and below the bar, about 9 cm away. It is further to the side of and below the faucet, about 76 cm away.", "target": "soap dish", "split": "train"}
{"id": "scene0702_00_3", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"bathtub\", \"references\": [{\"object\": \"bathtub\", \"relation\": \"side_above\", \"distance_cm\": 6.1}, {\"object\": \"bar\", \"relation\": \"side_below\", \"distance_cm\": 9.3}]}", "target": "soap dish", "split": "train"}
{"id": "scene0702_00_4", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "train"}
{"id": "scene0702_00_5", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the telephone in this scene?\nA. toilet paper\nB. trash can\nC. toilet\nD. bar", "choices": ["A. toilet paper", "B. trash can", "C. toilet", "D. bar"], "answer": "A. toilet paper", "target": "telephone", "split": "train"}
{"id": "scene0702_00_6", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is to the side of and above the toilet paper, about 16 cm away. It is also to the side of and above the toilet, about 50 cm away. It is further to the side of and above the trash can, about 67 cm away.", "target": "telephone", "split": "train"}
{"id": "scene0702_00_7", "scene_id": "scene0702_00", "source_dataset": "scannet200", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"side_above\", \"distance_cm\": 16.4}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 50.3}]}", "target": "telephone", "split": "train"}
{"id": "scene0705_00_0", "scene_id": "scene0705_00", "source_dataset": "scannet200", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "scene0705_00_1", "scene_id": "scene0705_00", "source_dataset": "scannet200", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. desk\nB. chair\nC. monitor\nD. table", "choices": ["A. desk", "B. chair", "C. monitor", "D. table"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "00a231a370_0", "scene_id": "00a231a370", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "00a231a370_1", "scene_id": "00a231a370", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the cloth, about 3 cm away. It is also to the side of and above the table, about 3 cm away. It is further to the side of and below the mirror frame, about 7 cm away.", "target": "bottle", "split": "train"}
{"id": "00a231a370_2", "scene_id": "00a231a370", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cloth\", \"relation\": \"side_above\", \"distance_cm\": 3.1}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 3.2}]}", "target": "bottle", "split": "train"}
{"id": "00a231a370_3", "scene_id": "00a231a370", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "train"}
{"id": "00a231a370_4", "scene_id": "00a231a370", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the purse in this scene?\nA. guitar stand\nB. mirror frame\nC. chair\nD. cloth", "choices": ["A. guitar stand", "B. mirror frame", "C. chair", "D. cloth"], "answer": "B. mirror frame", "target": "purse", "split": "train"}
{"id": "00a231a370_5", "scene_id": "00a231a370", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the purse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The purse is on the table. It is also to the side of and below the mirror frame, about 13 cm away. It is further to the side of and above the cloth, about 16 cm away.", "target": "purse", "split": "train"}
{"id": "00a231a370_6", "scene_id": "00a231a370", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the purse in structured JSON format.", "choices": null, "answer": "{\"target\": \"purse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mirror frame\", \"relation\": \"side_below\", \"distance_cm\": 12.9}, {\"object\": \"cloth\", \"relation\": \"side_above\", \"distance_cm\": 15.6}]}", "target": "purse", "split": "train"}
{"id": "01ce24e652_0", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "01ce24e652_1", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphone case in this scene?\nA. jacket\nB. keyboard\nC. laptop\nD. chair", "choices": ["A. jacket", "B. keyboard", "C. laptop", "D. chair"], "answer": "C. laptop", "target": "headphone case", "split": "train"}
{"id": "01ce24e652_2", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphone case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphone case is on the table. It is also next to the laptop, about 10 cm away. It is further next to the keyboard, about 13 cm away.", "target": "headphone case", "split": "train"}
{"id": "01ce24e652_3", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphone case in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphone case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 10.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 13.0}]}", "target": "headphone case", "split": "train"}
{"id": "01ce24e652_4", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "01ce24e652_5", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. jacket\nB. paper bag\nC. mouse\nD. headphone case", "choices": ["A. jacket", "B. paper bag", "C. mouse", "D. headphone case"], "answer": "A. jacket", "target": "keyboard", "split": "train"}
{"id": "01ce24e652_6", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the jacket, about 5 cm away. It is also next to the paper bag, about 10 cm away. It is further next to the mouse, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "01ce24e652_7", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 5.2}, {\"object\": \"paper bag\", \"relation\": \"next_to\", \"distance_cm\": 10.2}]}", "target": "keyboard", "split": "train"}
{"id": "01ce24e652_8", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "01ce24e652_9", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. laptop\nB. jacket\nC. keyboard\nD. chair", "choices": ["A. laptop", "B. jacket", "C. keyboard", "D. chair"], "answer": "A. laptop", "target": "mouse", "split": "train"}
{"id": "01ce24e652_10", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the laptop, about 8 cm away. It is also to the side of and above the chair, about 10 cm away. It is further next to the keyboard, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "01ce24e652_11", "scene_id": "01ce24e652", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 7.8}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 10.3}]}", "target": "mouse", "split": "train"}
{"id": "02c2ddee2a_0", "scene_id": "02c2ddee2a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "02c2ddee2a_1", "scene_id": "02c2ddee2a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. soap dispenser\nB. bathroom shelf\nC. split\nD. towel holder", "choices": ["A. soap dispenser", "B. bathroom shelf", "C. split", "D. towel holder"], "answer": "B. bathroom shelf", "target": "toothbrush", "split": "train"}
{"id": "02c2ddee2a_2", "scene_id": "02c2ddee2a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is to the side of and below the bathroom shelf, about 11 cm away. It is also to the side of and above the soap dispenser, about 21 cm away. It is further to the side of and above the towel holder, about 22 cm away.", "target": "toothbrush", "split": "train"}
{"id": "02c2ddee2a_3", "scene_id": "02c2ddee2a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"bathroom shelf\", \"references\": [{\"object\": \"bathroom shelf\", \"relation\": \"side_below\", \"distance_cm\": 10.8}, {\"object\": \"soap dispenser\", \"relation\": \"side_above\", \"distance_cm\": 20.8}]}", "target": "toothbrush", "split": "train"}
{"id": "2ebbca3e57_0", "scene_id": "2ebbca3e57", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "2ebbca3e57_1", "scene_id": "2ebbca3e57", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. table\nB. power socket\nC. monitor\nD. light switch", "choices": ["A. table", "B. power socket", "C. monitor", "D. light switch"], "answer": "D. light switch", "target": "phone charger", "split": "train"}
{"id": "2ebbca3e57_2", "scene_id": "2ebbca3e57", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the door, about 10 cm away. It is also below the light switch, about 75 cm away.", "target": "phone charger", "split": "train"}
{"id": "2ebbca3e57_3", "scene_id": "2ebbca3e57", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 9.9}, {\"object\": \"light switch\", \"relation\": \"below\", \"distance_cm\": 75.1}]}", "target": "phone charger", "split": "train"}
{"id": "2f5996ff01_0", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "2f5996ff01_1", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. umbrella\nB. mouse\nC. notebooks\nD. cloth", "choices": ["A. umbrella", "B. mouse", "C. notebooks", "D. cloth"], "answer": "A. umbrella", "target": "bottle", "split": "train"}
{"id": "2f5996ff01_2", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the desk. It is also to the side of and above the umbrella, about 4 cm away. It is further to the side of and above the notebooks, about 19 cm away.", "target": "bottle", "split": "train"}
{"id": "2f5996ff01_3", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"umbrella\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"notebooks\", \"relation\": \"side_above\", \"distance_cm\": 18.7}]}", "target": "bottle", "split": "train"}
{"id": "2f5996ff01_4", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "2f5996ff01_5", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. laptop charger\nB. notebooks\nC. spectacle case\nD. mouse", "choices": ["A. laptop charger", "B. notebooks", "C. spectacle case", "D. mouse"], "answer": "A. laptop charger", "target": "headphones", "split": "train"}
{"id": "2f5996ff01_6", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the desk. It is also next to the laptop charger, about 7 cm away. It is further next to the mouse, about 9 cm away.", "target": "headphones", "split": "train"}
{"id": "2f5996ff01_7", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop charger\", \"relation\": \"next_to\", \"distance_cm\": 7.0}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 9.5}]}", "target": "headphones", "split": "train"}
{"id": "2f5996ff01_8", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "2f5996ff01_9", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. mouse\nB. router\nC. spectacle case\nD. headphones", "choices": ["A. mouse", "B. router", "C. spectacle case", "D. headphones"], "answer": "C. spectacle case", "target": "laptop charger", "split": "train"}
{"id": "2f5996ff01_10", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is on the desk. It is also next to the spectacle case, about 5 cm away. It is further next to the headphones, about 7 cm away.", "target": "laptop charger", "split": "train"}
{"id": "2f5996ff01_11", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"spectacle case\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 7.0}]}", "target": "laptop charger", "split": "train"}
{"id": "2f5996ff01_12", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "No", "target": "phone charger", "split": "train"}
{"id": "2f5996ff01_13", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. notebooks\nB. headphones\nC. cloth\nD. laptop", "choices": ["A. notebooks", "B. headphones", "C. cloth", "D. laptop"], "answer": "D. laptop", "target": "mouse", "split": "train"}
{"id": "2f5996ff01_14", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the laptop, about 7 cm away. It is further next to the headphones, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "2f5996ff01_15", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 7.4}, {\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 9.5}]}", "target": "mouse", "split": "train"}
{"id": "2f5996ff01_16", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "2f5996ff01_17", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. plastic bag\nB. heater\nC. laptop\nD. router", "choices": ["A. plastic bag", "B. heater", "C. laptop", "D. router"], "answer": "D. router", "target": "power strip", "split": "train"}
{"id": "2f5996ff01_18", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the desk. It is also next to the router, about 4 cm away. It is further to the side of and above the heater, about 16 cm away.", "target": "power strip", "split": "train"}
{"id": "2f5996ff01_19", "scene_id": "2f5996ff01", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"router\", \"relation\": \"next_to\", \"distance_cm\": 4.1}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 15.8}]}", "target": "power strip", "split": "train"}
{"id": "2f6f83ea1f_0", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "2f6f83ea1f_1", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. bottle\nB. jar\nC. power socket\nD. power switch", "choices": ["A. bottle", "B. jar", "C. power socket", "D. power switch"], "answer": "B. jar", "target": "cup", "split": "train"}
{"id": "2f6f83ea1f_2", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is near the jar, about 16 cm away. It is also above the bottle, about 29 cm away. It is further to the side of and above the power socket, about 33 cm away.", "target": "cup", "split": "train"}
{"id": "2f6f83ea1f_3", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"jar\", \"relation\": \"near\", \"distance_cm\": 15.5}, {\"object\": \"bottle\", \"relation\": \"above\", \"distance_cm\": 28.5}]}", "target": "cup", "split": "train"}
{"id": "2f6f83ea1f_4", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "2f6f83ea1f_5", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. book\nB. chair\nC. monitor\nD. mouse", "choices": ["A. book", "B. chair", "C. monitor", "D. mouse"], "answer": "B. chair", "target": "keyboard", "split": "train"}
{"id": "2f6f83ea1f_6", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the chair, about 4 cm away. It is further next to the mouse, about 6 cm away.", "target": "keyboard", "split": "train"}
{"id": "2f6f83ea1f_7", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 4.4}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.9}]}", "target": "keyboard", "split": "train"}
{"id": "2f6f83ea1f_8", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the deodorant bottle appear?", "choices": null, "answer": "No", "target": "deodorant bottle", "split": "train"}
{"id": "2f6f83ea1f_9", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. book\nC. chair\nD. keyboard", "choices": ["A. monitor", "B. book", "C. chair", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "2f6f83ea1f_10", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 6 cm away. It is further to the side of and below the monitor, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "2f6f83ea1f_11", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.1}]}", "target": "mouse", "split": "train"}
{"id": "2f6f83ea1f_12", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "2f6f83ea1f_13", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. box\nB. book\nC. chair\nD. monitor", "choices": ["A. box", "B. book", "C. chair", "D. monitor"], "answer": "B. book", "target": "power strip", "split": "train"}
{"id": "2f6f83ea1f_14", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also next to the book, about 12 cm away. It is further to the side of and above the chair, about 37 cm away.", "target": "power strip", "split": "train"}
{"id": "2f6f83ea1f_15", "scene_id": "2f6f83ea1f", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 12.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 36.7}]}", "target": "power strip", "split": "train"}
{"id": "302a7f6b67_0", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "302a7f6b67_1", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. socket\nB. monitor\nC. keyboard\nD. telephone", "choices": ["A. socket", "B. monitor", "C. keyboard", "D. telephone"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "302a7f6b67_2", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 6 cm away. It is further to the side of and below the monitor, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "302a7f6b67_3", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.0}]}", "target": "mouse", "split": "train"}
{"id": "302a7f6b67_4", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "302a7f6b67_5", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. keyboard\nB. stamp\nC. monitor\nD. paper", "choices": ["A. keyboard", "B. stamp", "C. monitor", "D. paper"], "answer": "C. monitor", "target": "stapler", "split": "train"}
{"id": "302a7f6b67_6", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is on the table. It is also to the side of and below the monitor, about 14 cm away. It is further near the stamp, about 20 cm away.", "target": "stapler", "split": "train"}
{"id": "302a7f6b67_7", "scene_id": "302a7f6b67", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.6}, {\"object\": \"stamp\", \"relation\": \"near\", \"distance_cm\": 19.9}]}", "target": "stapler", "split": "train"}
{"id": "307e3262f1_0", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "307e3262f1_1", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bedside table\nB. book\nC. bedside lamp\nD. chair", "choices": ["A. bedside table", "B. book", "C. bedside lamp", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "train"}
{"id": "307e3262f1_2", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 24 cm away.", "target": "backpack", "split": "train"}
{"id": "307e3262f1_3", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 24.1}]}", "target": "backpack", "split": "train"}
{"id": "307e3262f1_4", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "307e3262f1_5", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. headboard\nB. box\nC. bedside table\nD. power socket", "choices": ["A. headboard", "B. box", "C. bedside table", "D. power socket"], "answer": "B. box", "target": "power adapter", "split": "train"}
{"id": "307e3262f1_6", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and above the box, about 6 cm away. It is also to the side of and below the headboard, about 12 cm away. It is further to the side of and below the bedside table, about 27 cm away.", "target": "power adapter", "split": "train"}
{"id": "307e3262f1_7", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 6.0}, {\"object\": \"headboard\", \"relation\": \"side_below\", \"distance_cm\": 12.1}]}", "target": "power adapter", "split": "train"}
{"id": "307e3262f1_8", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "307e3262f1_9", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. power sockets\nB. mirror\nC. tv\nD. book", "choices": ["A. power sockets", "B. mirror", "C. tv", "D. book"], "answer": "C. tv", "target": "tv remote", "split": "train"}
{"id": "307e3262f1_10", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is under the tv, about 4 cm away. It is also to the side of and above the mirror, about 47 cm away. It is further to the side of and above the power sockets, about 73 cm away.", "target": "tv remote", "split": "train"}
{"id": "307e3262f1_11", "scene_id": "307e3262f1", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"tv cabinet\", \"references\": [{\"object\": \"tv\", \"relation\": \"under\", \"distance_cm\": 4.3}, {\"object\": \"mirror\", \"relation\": \"side_above\", \"distance_cm\": 47.3}]}", "target": "tv remote", "split": "train"}
{"id": "30f4a2b44d_0", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "30f4a2b44d_1", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. split\nB. bin\nC. box\nD. whiteboard eraser", "choices": ["A. split", "B. bin", "C. box", "D. whiteboard eraser"], "answer": "D. whiteboard eraser", "target": "bottle", "split": "train"}
{"id": "30f4a2b44d_2", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also to the side of and above the whiteboard eraser, about 7 cm away. It is further to the side of and above the split, about 17 cm away.", "target": "bottle", "split": "train"}
{"id": "30f4a2b44d_3", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"whiteboard eraser\", \"relation\": \"side_above\", \"distance_cm\": 6.9}, {\"object\": \"split\", \"relation\": \"side_above\", \"distance_cm\": 16.6}]}", "target": "bottle", "split": "train"}
{"id": "30f4a2b44d_4", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "Yes", "target": "charger", "split": "train"}
{"id": "30f4a2b44d_5", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. keyboard\nB. power socket\nC. bin\nD. telephone", "choices": ["A. keyboard", "B. power socket", "C. bin", "D. telephone"], "answer": "B. power socket", "target": "charger", "split": "train"}
{"id": "30f4a2b44d_6", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is to the side of and below the power socket, about 68 cm away. It is also to the side of and below the keyboard, about 75 cm away. It is further to the side of and below the telephone, about 81 cm away.", "target": "charger", "split": "train"}
{"id": "30f4a2b44d_7", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"power socket\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 67.9}, {\"object\": \"keyboard\", \"relation\": \"side_below\", \"distance_cm\": 75.5}]}", "target": "charger", "split": "train"}
{"id": "30f4a2b44d_8", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "30f4a2b44d_9", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. papers\nB. mouse\nC. chair\nD. cabinet", "choices": ["A. papers", "B. mouse", "C. chair", "D. cabinet"], "answer": "A. papers", "target": "cup", "split": "train"}
{"id": "30f4a2b44d_10", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and above the papers, about 3 cm away. It is further to the side of and above the cabinet, about 21 cm away.", "target": "cup", "split": "train"}
{"id": "30f4a2b44d_11", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"papers\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 21.3}]}", "target": "cup", "split": "train"}
{"id": "30f4a2b44d_12", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "Yes", "target": "purse", "split": "train"}
{"id": "30f4a2b44d_13", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the purse in this scene?\nA. picture\nB. split\nC. books\nD. decoration", "choices": ["A. picture", "B. split", "C. books", "D. decoration"], "answer": "D. decoration", "target": "purse", "split": "train"}
{"id": "30f4a2b44d_14", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the purse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The purse is on the bookshelf. It is also to the side of and above the decoration, about 18 cm away. It is further to the side of and below the books, about 29 cm away.", "target": "purse", "split": "train"}
{"id": "30f4a2b44d_15", "scene_id": "30f4a2b44d", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the purse in structured JSON format.", "choices": null, "answer": "{\"target\": \"purse\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"decoration\", \"relation\": \"side_above\", \"distance_cm\": 18.0}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 29.0}]}", "target": "purse", "split": "train"}
{"id": "3134f5f3d6_0", "scene_id": "3134f5f3d6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "3134f5f3d6_1", "scene_id": "3134f5f3d6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "3134f5f3d6_2", "scene_id": "3134f5f3d6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "Yes", "target": "remote controller", "split": "train"}
{"id": "3134f5f3d6_3", "scene_id": "3134f5f3d6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote controller in this scene?\nA. chair\nB. marker\nC. cabinet\nD. cables", "choices": ["A. chair", "B. marker", "C. cabinet", "D. cables"], "answer": "A. chair", "target": "remote controller", "split": "train"}
{"id": "3134f5f3d6_4", "scene_id": "3134f5f3d6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote controller in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote controller is on the table. It is also to the side of and above the chair, about 27 cm away. It is further to the side of and above the cabinet, about 60 cm away.", "target": "remote controller", "split": "train"}
{"id": "3134f5f3d6_5", "scene_id": "3134f5f3d6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote controller in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote controller\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 27.3}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 60.4}]}", "target": "remote controller", "split": "train"}
{"id": "31a2c91c43_0", "scene_id": "31a2c91c43", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the deodorant bottle appear?", "choices": null, "answer": "No", "target": "deodorant bottle", "split": "train"}
{"id": "31a2c91c43_1", "scene_id": "31a2c91c43", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is to the side of and above the sink, about 5 cm away. It is also above the toothbrush holder, about 5 cm away. It is further next to the shampoo bottle, about 6 cm away.", "target": "toothbrush", "split": "train"}
{"id": "31a2c91c43_2", "scene_id": "31a2c91c43", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"tooth paste\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 5.4}, {\"object\": \"toothbrush holder\", \"relation\": \"above\", \"distance_cm\": 5.4}]}", "target": "toothbrush", "split": "train"}
{"id": "320c3af000_0", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "320c3af000_1", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. cabinet\nC. keyboard\nD. telephone", "choices": ["A. monitor", "B. cabinet", "C. keyboard", "D. telephone"], "answer": "D. telephone", "target": "bottle", "split": "train"}
{"id": "320c3af000_2", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the telephone, about 13 cm away. It is further to the side of and below the monitor, about 25 cm away.", "target": "bottle", "split": "train"}
{"id": "320c3af000_3", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"telephone\", \"relation\": \"side_above\", \"distance_cm\": 12.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 24.9}]}", "target": "bottle", "split": "train"}
{"id": "320c3af000_4", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "Yes", "target": "headset", "split": "train"}
{"id": "320c3af000_5", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headset in this scene?\nA. mouse\nB. keyboard\nC. monitor\nD. heater", "choices": ["A. mouse", "B. keyboard", "C. monitor", "D. heater"], "answer": "D. heater", "target": "headset", "split": "train"}
{"id": "320c3af000_6", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is to the side of and above the heater, about 19 cm away. It is also near the mouse, about 31 cm away. It is further near the keyboard, about 38 cm away.", "target": "headset", "split": "train"}
{"id": "320c3af000_7", "scene_id": "320c3af000", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 19.1}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 30.7}]}", "target": "headset", "split": "train"}
{"id": "32280ecbca_0", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "32280ecbca_1", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also near the tv, about 39 cm away. It is further to the side of and above the banana, about 45 cm away.", "target": "bottle", "split": "train"}
{"id": "32280ecbca_2", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"tv\", \"relation\": \"near\", \"distance_cm\": 39.2}, {\"object\": \"banana\", \"relation\": \"side_above\", \"distance_cm\": 45.3}]}", "target": "bottle", "split": "train"}
{"id": "32280ecbca_3", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "Yes", "target": "charger", "split": "train"}
{"id": "32280ecbca_4", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. chair\nB. lamp\nC. trash can\nD. laptop", "choices": ["A. chair", "B. lamp", "C. trash can", "D. laptop"], "answer": "D. laptop", "target": "charger", "split": "train"}
{"id": "32280ecbca_5", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is on the table. It is also next to the laptop, about 10 cm away. It is further to the side of and above the chair, about 49 cm away.", "target": "charger", "split": "train"}
{"id": "32280ecbca_6", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 9.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 49.3}]}", "target": "charger", "split": "train"}
{"id": "32280ecbca_7", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the deodorant bottle appear?", "choices": null, "answer": "No", "target": "deodorant bottle", "split": "train"}
{"id": "32280ecbca_8", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. lamp\nB. banana\nC. bottle\nD. tv", "choices": ["A. lamp", "B. banana", "C. bottle", "D. tv"], "answer": "D. tv", "target": "remote", "split": "train"}
{"id": "32280ecbca_9", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is on the shelf. It is also to the side of and below the tv, about 23 cm away. It is further near the banana, about 35 cm away.", "target": "remote", "split": "train"}
{"id": "32280ecbca_10", "scene_id": "32280ecbca", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 23.2}, {\"object\": \"banana\", \"relation\": \"near\", \"distance_cm\": 35.0}]}", "target": "remote", "split": "train"}
{"id": "324d07a5b3_0", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "Yes", "target": "glasses case", "split": "train"}
{"id": "324d07a5b3_1", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. sticky notes\nB. bag\nC. chair\nD. shelf", "choices": ["A. sticky notes", "B. bag", "C. chair", "D. shelf"], "answer": "B. bag", "target": "glasses case", "split": "train"}
{"id": "324d07a5b3_2", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is on the table. It is also near the bag, about 29 cm away. It is further to the side of and below the sticky notes, about 32 cm away.", "target": "glasses case", "split": "train"}
{"id": "324d07a5b3_3", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bag\", \"relation\": \"near\", \"distance_cm\": 28.9}, {\"object\": \"sticky notes\", \"relation\": \"side_below\", \"distance_cm\": 32.1}]}", "target": "glasses case", "split": "train"}
{"id": "324d07a5b3_4", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "train"}
{"id": "324d07a5b3_5", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. wlan router\nB. sticky notes\nC. table\nD. book", "choices": ["A. wlan router", "B. sticky notes", "C. table", "D. book"], "answer": "A. wlan router", "target": "pen holder", "split": "train"}
{"id": "324d07a5b3_6", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is next to the wlan router, about 5 cm away. It is also to the side of and above the table, about 7 cm away. It is further to the side of and below the sticky notes, about 14 cm away.", "target": "pen holder", "split": "train"}
{"id": "324d07a5b3_7", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"wlan router\", \"relation\": \"next_to\", \"distance_cm\": 5.4}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 6.9}]}", "target": "pen holder", "split": "train"}
{"id": "324d07a5b3_8", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "324d07a5b3_9", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and above the jar, about 3 cm away. It is also next to the bottle, about 4 cm away. It is further to the side of and above the split, about 9 cm away.", "target": "spray bottle", "split": "train"}
{"id": "324d07a5b3_10", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"jar\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 4.1}]}", "target": "spray bottle", "split": "train"}
{"id": "324d07a5b3_11", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "Yes", "target": "toothbrush", "split": "train"}
{"id": "324d07a5b3_12", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. jar\nB. bottle\nC. split\nD. soap", "choices": ["A. jar", "B. bottle", "C. split", "D. soap"], "answer": "C. split", "target": "toothbrush", "split": "train"}
{"id": "324d07a5b3_13", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is next to the split, about 4 cm away. It is also above the jar, about 7 cm away. It is further below the bottle, about 11 cm away.", "target": "toothbrush", "split": "train"}
{"id": "324d07a5b3_14", "scene_id": "324d07a5b3", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"split\", \"relation\": \"next_to\", \"distance_cm\": 3.6}, {\"object\": \"jar\", \"relation\": \"above\", \"distance_cm\": 7.1}]}", "target": "toothbrush", "split": "train"}
{"id": "3423e509af_0", "scene_id": "3423e509af", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "3423e509af_1", "scene_id": "3423e509af", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also next to the tv remote, about 10 cm away. It is further to the side of and below the tv, about 10 cm away.", "target": "bottle", "split": "train"}
{"id": "3423e509af_2", "scene_id": "3423e509af", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"tv remote\", \"relation\": \"next_to\", \"distance_cm\": 9.5}, {\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 9.6}]}", "target": "bottle", "split": "train"}
{"id": "3423e509af_3", "scene_id": "3423e509af", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "Yes", "target": "tv remote", "split": "train"}
{"id": "3423e509af_4", "scene_id": "3423e509af", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. clothes\nB. bag\nC. bottle\nD. tv", "choices": ["A. clothes", "B. bag", "C. bottle", "D. tv"], "answer": "C. bottle", "target": "tv remote", "split": "train"}
{"id": "3423e509af_5", "scene_id": "3423e509af", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the shelf. It is also next to the bottle, about 10 cm away. It is further next to the bag, about 14 cm away.", "target": "tv remote", "split": "train"}
{"id": "3423e509af_6", "scene_id": "3423e509af", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 9.5}, {\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 14.3}]}", "target": "tv remote", "split": "train"}
{"id": "35050f41c5_0", "scene_id": "35050f41c5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "35050f41c5_1", "scene_id": "35050f41c5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. screen\nB. cardboard\nC. eraser\nD. blackboard", "choices": ["A. screen", "B. cardboard", "C. eraser", "D. blackboard"], "answer": "A. screen", "target": "cable", "split": "train"}
{"id": "35050f41c5_2", "scene_id": "35050f41c5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and below the screen, about 27 cm away. It is also to the side of and below the cardboard, about 31 cm away.", "target": "cable", "split": "train"}
{"id": "35050f41c5_3", "scene_id": "35050f41c5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"stand\", \"references\": [{\"object\": \"screen\", \"relation\": \"side_below\", \"distance_cm\": 27.3}, {\"object\": \"cardboard\", \"relation\": \"side_below\", \"distance_cm\": 30.9}]}", "target": "cable", "split": "train"}
{"id": "35050f41c5_4", "scene_id": "35050f41c5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "355e5e32db_0", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "355e5e32db_1", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. partition\nB. monitor\nC. office chair\nD. mouse", "choices": ["A. partition", "B. monitor", "C. office chair", "D. mouse"], "answer": "B. monitor", "target": "keyboard", "split": "train"}
{"id": "355e5e32db_2", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the monitor, about 3 cm away. It is further next to the mouse, about 5 cm away.", "target": "keyboard", "split": "train"}
{"id": "355e5e32db_3", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.0}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.4}]}", "target": "keyboard", "split": "train"}
{"id": "355e5e32db_4", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "355e5e32db_5", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. cabinet\nC. keyboard\nD. monitor", "choices": ["A. office chair", "B. cabinet", "C. keyboard", "D. monitor"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "355e5e32db_6", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 5 cm away. It is further to the side of and above the office chair, about 12 cm away.", "target": "mouse", "split": "train"}
{"id": "355e5e32db_7", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.0}]}", "target": "mouse", "split": "train"}
{"id": "355e5e32db_8", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "355e5e32db_9", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. power strip\nB. desk\nC. monitor\nD. office chair", "choices": ["A. power strip", "B. desk", "C. monitor", "D. office chair"], "answer": "C. monitor", "target": "mug", "split": "train"}
{"id": "355e5e32db_10", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and below the monitor, about 7 cm away. It is further to the side of and above the office chair, about 28 cm away.", "target": "mug", "split": "train"}
{"id": "355e5e32db_11", "scene_id": "355e5e32db", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.2}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 28.3}]}", "target": "mug", "split": "train"}
{"id": "364f01bc18_0", "scene_id": "364f01bc18", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "364f01bc18_1", "scene_id": "364f01bc18", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. monitor\nB. table\nC. mouse\nD. keyboard", "choices": ["A. monitor", "B. table", "C. mouse", "D. keyboard"], "answer": "B. table", "target": "extension cord", "split": "train"}
{"id": "364f01bc18_2", "scene_id": "364f01bc18", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the table.", "target": "extension cord", "split": "train"}
{"id": "364f01bc18_3", "scene_id": "364f01bc18", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"table\", \"references\": []}", "target": "extension cord", "split": "train"}
{"id": "373ac8fbdd_0", "scene_id": "373ac8fbdd", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "373ac8fbdd_1", "scene_id": "373ac8fbdd", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. pillow\nB. blanket\nC. bottle\nD. table", "choices": ["A. pillow", "B. blanket", "C. bottle", "D. table"], "answer": "D. table", "target": "remote", "split": "train"}
{"id": "373ac8fbdd_2", "scene_id": "373ac8fbdd", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is on the sofa. It is also near the blanket, about 35 cm away. It is further to the side of and above the carpet, about 41 cm away.", "target": "remote", "split": "train"}
{"id": "373ac8fbdd_3", "scene_id": "373ac8fbdd", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"blanket\", \"relation\": \"near\", \"distance_cm\": 35.0}, {\"object\": \"carpet\", \"relation\": \"side_above\", \"distance_cm\": 40.6}]}", "target": "remote", "split": "train"}
{"id": "37562e7f48_0", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "37562e7f48_1", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. office chair\nB. mouse pad\nC. clothes\nD. table", "choices": ["A. office chair", "B. mouse pad", "C. clothes", "D. table"], "answer": "D. table", "target": "headphones", "split": "train"}
{"id": "37562e7f48_2", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the cabinet. It is also to the side of and above the table, about 14 cm away. It is further near the mouse pad, about 29 cm away.", "target": "headphones", "split": "train"}
{"id": "37562e7f48_3", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 14.3}, {\"object\": \"mouse pad\", \"relation\": \"near\", \"distance_cm\": 29.1}]}", "target": "headphones", "split": "train"}
{"id": "37562e7f48_4", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "No", "target": "soap", "split": "train"}
{"id": "37562e7f48_5", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. clothes\nB. office chair\nC. table fan\nD. monitor", "choices": ["A. clothes", "B. office chair", "C. table fan", "D. monitor"], "answer": "C. table fan", "target": "keyboard", "split": "train"}
{"id": "37562e7f48_6", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the table fan, about 16 cm away. It is further to the side of and above the clothes, about 25 cm away.", "target": "keyboard", "split": "train"}
{"id": "37562e7f48_7", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table fan\", \"relation\": \"side_below\", \"distance_cm\": 16.2}, {\"object\": \"clothes\", \"relation\": \"side_above\", \"distance_cm\": 24.6}]}", "target": "keyboard", "split": "train"}
{"id": "37562e7f48_8", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "Yes", "target": "mouse pad", "split": "train"}
{"id": "37562e7f48_9", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is to the side of and above the chair, about 13 cm away. It is also to the side of and above the bag, about 14 cm away. It is further to the side of and above the cabinet, about 15 cm away.", "target": "mouse pad", "split": "train"}
{"id": "37562e7f48_10", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 13.5}, {\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 13.8}]}", "target": "mouse pad", "split": "train"}
{"id": "37562e7f48_11", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "37562e7f48_12", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. books\nB. light switch\nC. tray\nD. shelf", "choices": ["A. books", "B. light switch", "C. tray", "D. shelf"], "answer": "B. light switch", "target": "power strip", "split": "train"}
{"id": "37562e7f48_13", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the shelf. It is also next to the light switch, about 5 cm away. It is further to the side of and above the tray, about 21 cm away.", "target": "power strip", "split": "train"}
{"id": "37562e7f48_14", "scene_id": "37562e7f48", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"light switch\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"tray\", \"relation\": \"side_above\", \"distance_cm\": 20.6}]}", "target": "power strip", "split": "train"}
{"id": "37c9538a2b_0", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "37c9538a2b_1", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. table\nB. bluetooth speaker\nC. chair\nD. cable duct", "choices": ["A. table", "B. bluetooth speaker", "C. chair", "D. cable duct"], "answer": "A. table", "target": "laptop charger", "split": "train"}
{"id": "37c9538a2b_2", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the table, about 19 cm away. It is also to the side of and below the chair, about 27 cm away. It is further to the side of and below the cable duct, about 30 cm away.", "target": "laptop charger", "split": "train"}
{"id": "37c9538a2b_3", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 19.5}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 26.9}]}", "target": "laptop charger", "split": "train"}
{"id": "37c9538a2b_4", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "train"}
{"id": "37c9538a2b_5", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. plant\nB. vase\nC. table lamp\nD. bluetooth speaker", "choices": ["A. plant", "B. vase", "C. table lamp", "D. bluetooth speaker"], "answer": "C. table lamp", "target": "tv remote", "split": "train"}
{"id": "37c9538a2b_6", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the table. It is also to the side of and below the table lamp, about 18 cm away. It is further to the side of and below the plant, about 21 cm away.", "target": "tv remote", "split": "train"}
{"id": "37c9538a2b_7", "scene_id": "37c9538a2b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 18.2}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 21.4}]}", "target": "tv remote", "split": "train"}
{"id": "37ea1c52f0_0", "scene_id": "37ea1c52f0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "37ea1c52f0_1", "scene_id": "37ea1c52f0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. office chair\nB. monitor\nC. joined tables\nD. table", "choices": ["A. office chair", "B. monitor", "C. joined tables", "D. table"], "answer": "A. office chair", "target": "power strip", "split": "train"}
{"id": "37ea1c52f0_2", "scene_id": "37ea1c52f0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and above the office chair, about 55 cm away.", "target": "power strip", "split": "train"}
{"id": "37ea1c52f0_3", "scene_id": "37ea1c52f0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 54.6}]}", "target": "power strip", "split": "train"}
{"id": "3878be46f0_0", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "3878be46f0_1", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. bathroom shelf\nB. bathroom cabinet\nC. towel radiator\nD. bathroom wall", "choices": ["A. bathroom shelf", "B. bathroom cabinet", "C. towel radiator", "D. bathroom wall"], "answer": "D. bathroom wall", "target": "bottle", "split": "train"}
{"id": "3878be46f0_2", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the bathroom cabinet, about 30 cm away. It is also to the side of and above the towel radiator, about 38 cm away. It is further to the side of and above the power socket, about 46 cm away.", "target": "bottle", "split": "train"}
{"id": "3878be46f0_3", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bathroom shelf\", \"references\": [{\"object\": \"bathroom cabinet\", \"relation\": \"side_below\", \"distance_cm\": 30.4}, {\"object\": \"towel radiator\", \"relation\": \"side_above\", \"distance_cm\": 37.9}]}", "target": "bottle", "split": "train"}
{"id": "3878be46f0_4", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "3878be46f0_5", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. washing machine\nB. toilet paper roll\nC. toilet\nD. box", "choices": ["A. washing machine", "B. toilet paper roll", "C. toilet", "D. box"], "answer": "C. toilet", "target": "cable", "split": "train"}
{"id": "3878be46f0_6", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and above the toilet, about 4 cm away. It is also to the side of and above the washing machine, about 5 cm away. It is further to the side of and below the box, about 10 cm away.", "target": "cable", "split": "train"}
{"id": "3878be46f0_7", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"toilet cabinet\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 4.1}, {\"object\": \"washing machine\", \"relation\": \"side_above\", \"distance_cm\": 5.1}]}", "target": "cable", "split": "train"}
{"id": "3878be46f0_8", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "3878be46f0_9", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap in this scene?\nA. bottle\nB. bathroom shelf\nC. bathroom wall\nD. faucet", "choices": ["A. bottle", "B. bathroom shelf", "C. bathroom wall", "D. faucet"], "answer": "C. bathroom wall", "target": "soap", "split": "train"}
{"id": "3878be46f0_10", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap is under the bathroom wall, about 4 cm away. It is also next to the faucet, about 7 cm away. It is further near the bottle, about 20 cm away.", "target": "soap", "split": "train"}
{"id": "3878be46f0_11", "scene_id": "3878be46f0", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"bathroom wall\", \"relation\": \"under\", \"distance_cm\": 3.6}, {\"object\": \"faucet\", \"relation\": \"next_to\", \"distance_cm\": 7.5}]}", "target": "soap", "split": "train"}
{"id": "38d58a7a31_0", "scene_id": "38d58a7a31", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "train"}
{"id": "38d58a7a31_1", "scene_id": "38d58a7a31", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. remote controller\nB. power socket\nC. table\nD. split", "choices": ["A. remote controller", "B. power socket", "C. table", "D. split"], "answer": "B. power socket", "target": "keyboard", "split": "train"}
{"id": "38d58a7a31_2", "scene_id": "38d58a7a31", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the power socket, about 17 cm away. It is also to the side of and above the door, about 18 cm away. It is further to the side of and above the table, about 23 cm away.", "target": "keyboard", "split": "train"}
{"id": "38d58a7a31_3", "scene_id": "38d58a7a31", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 16.9}, {\"object\": \"door\", \"relation\": \"side_above\", \"distance_cm\": 18.4}]}", "target": "keyboard", "split": "train"}
{"id": "38fcf02d0b_0", "scene_id": "38fcf02d0b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "38fcf02d0b_1", "scene_id": "38fcf02d0b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. jackets\nB. handbag\nC. bicycle helmet\nD. sturage basket", "choices": ["A. jackets", "B. handbag", "C. bicycle helmet", "D. sturage basket"], "answer": "A. jackets", "target": "headphones", "split": "train"}
{"id": "38fcf02d0b_2", "scene_id": "38fcf02d0b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the jackets, about 9 cm away. It is also to the side of and above the sturage basket, about 30 cm away. It is further to the side of and below the bicycle helmet, about 40 cm away.", "target": "headphones", "split": "train"}
{"id": "38fcf02d0b_3", "scene_id": "38fcf02d0b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"entryway shelf\", \"references\": [{\"object\": \"jackets\", \"relation\": \"side_above\", \"distance_cm\": 9.1}, {\"object\": \"sturage basket\", \"relation\": \"side_above\", \"distance_cm\": 29.5}]}", "target": "headphones", "split": "train"}
{"id": "390eda9157_0", "scene_id": "390eda9157", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "390eda9157_1", "scene_id": "390eda9157", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. heater\nB. coffee table\nC. whiteboard\nD. plant", "choices": ["A. heater", "B. coffee table", "C. whiteboard", "D. plant"], "answer": "B. coffee table", "target": "power adapter", "split": "train"}
{"id": "390eda9157_2", "scene_id": "390eda9157", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and below the coffee table, about 8 cm away. It is also to the side of and below the plant, about 38 cm away. It is further to the side of and below the heater, about 40 cm away.", "target": "power adapter", "split": "train"}
{"id": "390eda9157_3", "scene_id": "390eda9157", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"coffee table\", \"relation\": \"side_below\", \"distance_cm\": 8.2}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 38.1}]}", "target": "power adapter", "split": "train"}
{"id": "3928249b53_0", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "3928249b53_1", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also near the box, about 18 cm away. It is further near the mouse, about 18 cm away.", "target": "cup", "split": "train"}
{"id": "3928249b53_2", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"near\", \"distance_cm\": 18.1}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 18.2}]}", "target": "cup", "split": "train"}
{"id": "3928249b53_3", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "Yes", "target": "eraser", "split": "train"}
{"id": "3928249b53_4", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "3928249b53_5", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. monitor\nB. office chair\nC. mouse\nD. keyboard", "choices": ["A. monitor", "B. office chair", "C. mouse", "D. keyboard"], "answer": "D. keyboard", "target": "mug", "split": "train"}
{"id": "3928249b53_6", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also near the keyboard, about 20 cm away. It is further to the side of and above the office chair, about 28 cm away.", "target": "mug", "split": "train"}
{"id": "3928249b53_7", "scene_id": "3928249b53", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 19.7}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 27.8}]}", "target": "mug", "split": "train"}
{"id": "394a542a19_0", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "394a542a19_1", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. door frame\nC. mouse\nD. office chair", "choices": ["A. monitor", "B. door frame", "C. mouse", "D. office chair"], "answer": "C. mouse", "target": "bottle", "split": "train"}
{"id": "394a542a19_2", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the mouse, about 15 cm away. It is also to the side of and below the door frame, about 37 cm away. It is further to the side of and below the door, about 42 cm away.", "target": "bottle", "split": "train"}
{"id": "394a542a19_3", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"side_above\", \"distance_cm\": 14.8}, {\"object\": \"door frame\", \"relation\": \"side_below\", \"distance_cm\": 37.2}]}", "target": "bottle", "split": "train"}
{"id": "394a542a19_4", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "394a542a19_5", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. cup\nB. office table\nC. laptop\nD. office chair", "choices": ["A. cup", "B. office table", "C. laptop", "D. office chair"], "answer": "A. cup", "target": "pen", "split": "train"}
{"id": "394a542a19_6", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is next to the cup. It is also to the side of and above the office table, about 8 cm away. It is further to the side of and above the laptop, about 11 cm away.", "target": "pen", "split": "train"}
{"id": "394a542a19_7", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 2.2}, {\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 8.3}]}", "target": "pen", "split": "train"}
{"id": "394a542a19_8", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "394a542a19_9", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. laptop\nB. monitor\nC. headphones\nD. office chair", "choices": ["A. laptop", "B. monitor", "C. headphones", "D. office chair"], "answer": "A. laptop", "target": "pen holder", "split": "train"}
{"id": "394a542a19_10", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is next to the laptop, about 5 cm away. It is also to the side of and above the office chair, about 18 cm away. It is further near the headphones, about 29 cm away.", "target": "pen holder", "split": "train"}
{"id": "394a542a19_11", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 18.4}]}", "target": "pen holder", "split": "train"}
{"id": "394a542a19_12", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "394a542a19_13", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. office chair\nB. book\nC. cabinet\nD. keyboard", "choices": ["A. office chair", "B. book", "C. cabinet", "D. keyboard"], "answer": "D. keyboard", "target": "power strip", "split": "train"}
{"id": "394a542a19_14", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is near the keyboard, about 17 cm away. It is also to the side of and above the cabinet, about 21 cm away. It is further to the side of and above the book, about 47 cm away.", "target": "power strip", "split": "train"}
{"id": "394a542a19_15", "scene_id": "394a542a19", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 16.5}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 20.8}]}", "target": "power strip", "split": "train"}
{"id": "39e6ee46df_0", "scene_id": "39e6ee46df", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "39e6ee46df_1", "scene_id": "39e6ee46df", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and below the monitor, about 9 cm away. It is further next to the paper, about 9 cm away.", "target": "keyboard", "split": "train"}
{"id": "39e6ee46df_2", "scene_id": "39e6ee46df", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 8.9}]}", "target": "keyboard", "split": "train"}
{"id": "3a520af32a_0", "scene_id": "3a520af32a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "3a520af32a_1", "scene_id": "3a520af32a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. power socket\nB. faucet\nC. light switch\nD. sink", "choices": ["A. power socket", "B. faucet", "C. light switch", "D. sink"], "answer": "C. light switch", "target": "bottle", "split": "train"}
{"id": "3a520af32a_2", "scene_id": "3a520af32a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the light switch, about 3 cm away. It is also next to the faucet, about 11 cm away. It is further to the side of and below the power socket, about 29 cm away.", "target": "bottle", "split": "train"}
{"id": "3a520af32a_3", "scene_id": "3a520af32a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"light switch\", \"relation\": \"under\", \"distance_cm\": 3.4}, {\"object\": \"faucet\", \"relation\": \"next_to\", \"distance_cm\": 10.8}]}", "target": "bottle", "split": "train"}
{"id": "3b753d9ae8_0", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "3b753d9ae8_1", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. cable duct\nB. heater\nC. power strip\nD. chair", "choices": ["A. cable duct", "B. heater", "C. power strip", "D. chair"], "answer": "A. cable duct", "target": "cable", "split": "train"}
{"id": "3b753d9ae8_2", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and above the cable duct, about 35 cm away.", "target": "cable", "split": "train"}
{"id": "3b753d9ae8_3", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"cable duct\", \"references\": [{\"object\": \"cable duct\", \"relation\": \"side_above\", \"distance_cm\": 35.0}]}", "target": "cable", "split": "train"}
{"id": "3b753d9ae8_4", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "3b753d9ae8_5", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. heater\nB. cable\nC. cable duct\nD. chair", "choices": ["A. heater", "B. cable", "C. cable duct", "D. chair"], "answer": "D. chair", "target": "tv remote", "split": "train"}
{"id": "3b753d9ae8_6", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the table. It is also to the side of and above the chair, about 30 cm away.", "target": "tv remote", "split": "train"}
{"id": "3b753d9ae8_7", "scene_id": "3b753d9ae8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 29.5}]}", "target": "tv remote", "split": "train"}
{"id": "3cb9f85891_0", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "3cb9f85891_1", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. luggage\nB. bag\nC. dustpan and brush\nD. shoes", "choices": ["A. luggage", "B. bag", "C. dustpan and brush", "D. shoes"], "answer": "C. dustpan and brush", "target": "backpack", "split": "train"}
{"id": "3cb9f85891_2", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the dustpan and brush, about 9 cm away. It is also to the side of and above the shoes, about 51 cm away. It is further to the side of and below the bag, about 55 cm away.", "target": "backpack", "split": "train"}
{"id": "3cb9f85891_3", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"dustpan and brush\", \"relation\": \"side_below\", \"distance_cm\": 9.1}, {\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 51.3}]}", "target": "backpack", "split": "train"}
{"id": "3cb9f85891_4", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "3cb9f85891_5", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. bag\nB. dustpan and brush\nC. rice cooker\nD. backpack", "choices": ["A. bag", "B. dustpan and brush", "C. rice cooker", "D. backpack"], "answer": "C. rice cooker", "target": "phone charger", "split": "train"}
{"id": "3cb9f85891_6", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the rice cooker, about 62 cm away.", "target": "phone charger", "split": "train"}
{"id": "3cb9f85891_7", "scene_id": "3cb9f85891", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"rice cooker\", \"relation\": \"side_below\", \"distance_cm\": 62.0}]}", "target": "phone charger", "split": "train"}
{"id": "3ce6d36ab5_0", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "3ce6d36ab5_1", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glue bottle in this scene?\nA. floor lamp\nB. bread\nC. light switch\nD. wine bottle", "choices": ["A. floor lamp", "B. bread", "C. light switch", "D. wine bottle"], "answer": "B. bread", "target": "glue bottle", "split": "train"}
{"id": "3ce6d36ab5_2", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glue bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glue bottle is below the bread, about 20 cm away. It is also to the side of and above the floor lamp, about 22 cm away. It is further to the side of and below the wine bottle, about 30 cm away.", "target": "glue bottle", "split": "train"}
{"id": "3ce6d36ab5_3", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glue bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"glue bottle\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"bread\", \"relation\": \"below\", \"distance_cm\": 20.2}, {\"object\": \"floor lamp\", \"relation\": \"side_above\", \"distance_cm\": 22.3}]}", "target": "glue bottle", "split": "train"}
{"id": "3ce6d36ab5_4", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "3ce6d36ab5_5", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mat, about 5 cm away. It is also next to the laptop, about 7 cm away. It is further to the side of and above the office chair, about 14 cm away.", "target": "keyboard", "split": "train"}
{"id": "3ce6d36ab5_6", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mat\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "keyboard", "split": "train"}
{"id": "3ce6d36ab5_7", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "3ce6d36ab5_8", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. office chair\nB. shoebox\nC. table\nD. folding mattress", "choices": ["A. office chair", "B. shoebox", "C. table", "D. folding mattress"], "answer": "D. folding mattress", "target": "laptop charger", "split": "train"}
{"id": "3ce6d36ab5_9", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the folding mattress, about 11 cm away. It is also to the side of and above the shoebox, about 18 cm away. It is further to the side of and below the office chair, about 37 cm away.", "target": "laptop charger", "split": "train"}
{"id": "3ce6d36ab5_10", "scene_id": "3ce6d36ab5", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"folding mattress\", \"relation\": \"side_below\", \"distance_cm\": 10.5}, {\"object\": \"shoebox\", \"relation\": \"side_above\", \"distance_cm\": 17.8}]}", "target": "laptop charger", "split": "train"}
{"id": "3ce88cdad2_0", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "3ce88cdad2_1", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cloth\nB. basket\nC. vacuum cleaner\nD. tote bag", "choices": ["A. cloth", "B. basket", "C. vacuum cleaner", "D. tote bag"], "answer": "C. vacuum cleaner", "target": "backpack", "split": "train"}
{"id": "3ce88cdad2_2", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the vacuum cleaner, about 6 cm away. It is also to the side of and above the basket, about 14 cm away. It is further to the side of and above the tote bag, about 18 cm away.", "target": "backpack", "split": "train"}
{"id": "3ce88cdad2_3", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"vacuum cleaner\", \"relation\": \"side_above\", \"distance_cm\": 5.7}, {\"object\": \"basket\", \"relation\": \"side_above\", \"distance_cm\": 14.2}]}", "target": "backpack", "split": "train"}
{"id": "3ce88cdad2_4", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "3ce88cdad2_5", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. desk lamp\nC. cup\nD. mouse", "choices": ["A. monitor", "B. desk lamp", "C. cup", "D. mouse"], "answer": "B. desk lamp", "target": "keyboard", "split": "train"}
{"id": "3ce88cdad2_6", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the desk lamp, about 3 cm away. It is further next to the mouse, about 7 cm away.", "target": "keyboard", "split": "train"}
{"id": "3ce88cdad2_7", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk lamp\", \"relation\": \"under\", \"distance_cm\": 2.6}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 7.0}]}", "target": "keyboard", "split": "train"}
{"id": "3ce88cdad2_8", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "3ce88cdad2_9", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. keyboard\nC. desk lamp\nD. cup", "choices": ["A. office chair", "B. keyboard", "C. desk lamp", "D. cup"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "3ce88cdad2_10", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the keyboard, about 7 cm away. It is further to the side of and below the desk lamp, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "3ce88cdad2_11", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.0}, {\"object\": \"desk lamp\", \"relation\": \"side_below\", \"distance_cm\": 14.4}]}", "target": "mouse", "split": "train"}
{"id": "3ce88cdad2_12", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "3ce88cdad2_13", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. electric kettle\nB. shelf\nC. cloth\nD. wardrobe", "choices": ["A. electric kettle", "B. shelf", "C. cloth", "D. wardrobe"], "answer": "B. shelf", "target": "power strip", "split": "train"}
{"id": "3ce88cdad2_14", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the shelf. It is also to the side of and below the wardrobe, about 15 cm away. It is further to the side of and below the electric kettle, about 16 cm away.", "target": "power strip", "split": "train"}
{"id": "3ce88cdad2_15", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"wardrobe\", \"relation\": \"side_below\", \"distance_cm\": 14.6}, {\"object\": \"electric kettle\", \"relation\": \"side_below\", \"distance_cm\": 15.7}]}", "target": "power strip", "split": "train"}
{"id": "3ce88cdad2_16", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "No", "target": "notebook", "split": "train"}
{"id": "3ce88cdad2_17", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. sofa\nB. vacuum cleaner\nC. basket\nD. backpack", "choices": ["A. sofa", "B. vacuum cleaner", "C. basket", "D. backpack"], "answer": "C. basket", "target": "tote bag", "split": "train"}
{"id": "3ce88cdad2_18", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is next to the basket, about 5 cm away. It is also to the side of and below the sofa, about 13 cm away. It is further to the side of and below the backpack, about 18 cm away.", "target": "tote bag", "split": "train"}
{"id": "3ce88cdad2_19", "scene_id": "3ce88cdad2", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"basket\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"sofa\", \"relation\": \"side_below\", \"distance_cm\": 12.9}]}", "target": "tote bag", "split": "train"}
{"id": "3d0b7d4b4e_0", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "Yes", "target": "calculator", "split": "train"}
{"id": "3d0b7d4b4e_1", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the calculator in this scene?\nA. file binder\nB. book\nC. cardboard box\nD. documents", "choices": ["A. file binder", "B. book", "C. cardboard box", "D. documents"], "answer": "B. book", "target": "calculator", "split": "train"}
{"id": "3d0b7d4b4e_2", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the calculator in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The calculator is on the cabinet. It is also to the side of and below the documents, about 22 cm away. It is further to the side of and below the cardboard box, about 26 cm away.", "target": "calculator", "split": "train"}
{"id": "3d0b7d4b4e_3", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the calculator in structured JSON format.", "choices": null, "answer": "{\"target\": \"calculator\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"documents\", \"relation\": \"side_below\", \"distance_cm\": 22.3}, {\"object\": \"cardboard box\", \"relation\": \"side_below\", \"distance_cm\": 25.9}]}", "target": "calculator", "split": "train"}
{"id": "3d0b7d4b4e_4", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "3d0b7d4b4e_5", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. chair\nB. router\nC. desk\nD. power sockets", "choices": ["A. chair", "B. router", "C. desk", "D. power sockets"], "answer": "B. router", "target": "extension cord", "split": "train"}
{"id": "3d0b7d4b4e_6", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is near the router, about 25 cm away. It is also to the side of and below the desk, about 32 cm away. It is further to the side of and below the chair, about 34 cm away.", "target": "extension cord", "split": "train"}
{"id": "3d0b7d4b4e_7", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"router\", \"relation\": \"near\", \"distance_cm\": 25.1}, {\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 32.0}]}", "target": "extension cord", "split": "train"}
{"id": "3d0b7d4b4e_8", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "3d0b7d4b4e_9", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. documents\nB. laptop stand\nC. chair\nD. mouse", "choices": ["A. documents", "B. laptop stand", "C. chair", "D. mouse"], "answer": "B. laptop stand", "target": "keyboard", "split": "train"}
{"id": "3d0b7d4b4e_10", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the laptop stand, about 4 cm away. It is further next to the mouse, about 5 cm away.", "target": "keyboard", "split": "train"}
{"id": "3d0b7d4b4e_11", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop stand\", \"relation\": \"under\", \"distance_cm\": 3.6}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.2}]}", "target": "keyboard", "split": "train"}
{"id": "3d0b7d4b4e_12", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "3d0b7d4b4e_13", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. documents\nC. keyboard\nD. laptop charger", "choices": ["A. chair", "B. documents", "C. keyboard", "D. laptop charger"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "3d0b7d4b4e_14", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the keyboard, about 5 cm away. It is further next to the laptop charger, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "3d0b7d4b4e_15", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"laptop charger\", \"relation\": \"next_to\", \"distance_cm\": 10.0}]}", "target": "mouse", "split": "train"}
{"id": "3d0b7d4b4e_16", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "Yes", "target": "pencil case", "split": "train"}
{"id": "3d0b7d4b4e_17", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil case in this scene?\nA. chair\nB. paper bag\nC. stapler\nD. documents", "choices": ["A. chair", "B. paper bag", "C. stapler", "D. documents"], "answer": "D. documents", "target": "pencil case", "split": "train"}
{"id": "3d0b7d4b4e_18", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil case is on the cabinet. It is also next to the documents, about 6 cm away. It is further near the stapler, about 16 cm away.", "target": "pencil case", "split": "train"}
{"id": "3d0b7d4b4e_19", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"documents\", \"relation\": \"next_to\", \"distance_cm\": 5.8}, {\"object\": \"stapler\", \"relation\": \"near\", \"distance_cm\": 15.7}]}", "target": "pencil case", "split": "train"}
{"id": "3d0b7d4b4e_20", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "3d0b7d4b4e_21", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the post it note in this scene?\nA. window frame\nB. curtain rail\nC. curtain\nD. paper", "choices": ["A. window frame", "B. curtain rail", "C. curtain", "D. paper"], "answer": "D. paper", "target": "post it note", "split": "train"}
{"id": "3d0b7d4b4e_22", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the post it note in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The post it note is above the paper, about 15 cm away. It is also above the window, about 21 cm away. It is further to the side of and above the curtain, about 30 cm away.", "target": "post it note", "split": "train"}
{"id": "3d0b7d4b4e_23", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the post it note in structured JSON format.", "choices": null, "answer": "{\"target\": \"post it note\", \"support_surface\": \"window\", \"references\": [{\"object\": \"paper\", \"relation\": \"above\", \"distance_cm\": 14.9}, {\"object\": \"window\", \"relation\": \"above\", \"distance_cm\": 20.8}]}", "target": "post it note", "split": "train"}
{"id": "3d0b7d4b4e_24", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "3d0b7d4b4e_25", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is to the side of and above the cabinet, about 9 cm away. It is also to the side of and above the cardboard box, about 10 cm away. It is further to the side of and below the paper bag, about 15 cm away.", "target": "stapler", "split": "train"}
{"id": "3d0b7d4b4e_26", "scene_id": "3d0b7d4b4e", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 9.1}, {\"object\": \"cardboard box\", \"relation\": \"side_above\", \"distance_cm\": 10.1}]}", "target": "stapler", "split": "train"}
{"id": "3d838ee1ab_0", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "3d838ee1ab_1", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. clothes drying rack\nB. bottle\nC. chair\nD. desk lamp", "choices": ["A. clothes drying rack", "B. bottle", "C. chair", "D. desk lamp"], "answer": "C. chair", "target": "keyboard", "split": "train"}
{"id": "3d838ee1ab_2", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the chair, about 27 cm away. It is also to the side of and below the desk lamp, about 46 cm away. It is further to the side of and above the clothes drying rack, about 53 cm away.", "target": "keyboard", "split": "train"}
{"id": "3d838ee1ab_3", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 26.7}, {\"object\": \"desk lamp\", \"relation\": \"side_below\", \"distance_cm\": 46.0}]}", "target": "keyboard", "split": "train"}
{"id": "3d838ee1ab_4", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "train"}
{"id": "3d838ee1ab_5", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. chair\nB. table\nC. bottle\nD. trash can", "choices": ["A. chair", "B. table", "C. bottle", "D. trash can"], "answer": "B. table", "target": "power strip", "split": "train"}
{"id": "3d838ee1ab_6", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the table, about 17 cm away. It is also to the side of and below the chair, about 20 cm away. It is further to the side of and below the trash can, about 64 cm away.", "target": "power strip", "split": "train"}
{"id": "3d838ee1ab_7", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 16.9}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 19.8}]}", "target": "power strip", "split": "train"}
{"id": "3d838ee1ab_8", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "3d838ee1ab_9", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. sink\nB. kettle\nC. window frame\nD. frying pan", "choices": ["A. sink", "B. kettle", "C. window frame", "D. frying pan"], "answer": "D. frying pan", "target": "spray bottle", "split": "train"}
{"id": "3d838ee1ab_10", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is next to the frying pan, about 10 cm away. It is also next to the kettle, about 11 cm away. It is further to the side of and above the sink, about 14 cm away.", "target": "spray bottle", "split": "train"}
{"id": "3d838ee1ab_11", "scene_id": "3d838ee1ab", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"frying pan\", \"relation\": \"next_to\", \"distance_cm\": 9.5}, {\"object\": \"kettle\", \"relation\": \"next_to\", \"distance_cm\": 11.3}]}", "target": "spray bottle", "split": "train"}
{"id": "3e7e4b07c4_0", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "3e7e4b07c4_1", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothes drying rack\nB. messenger bag\nC. blanket\nD. bedside table", "choices": ["A. clothes drying rack", "B. messenger bag", "C. blanket", "D. bedside table"], "answer": "A. clothes drying rack", "target": "backpack", "split": "train"}
{"id": "3e7e4b07c4_2", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the clothes drying rack, about 8 cm away. It is also to the side of and below the bedside table, about 21 cm away. It is further to the side of and below the blanket, about 23 cm away.", "target": "backpack", "split": "train"}
{"id": "3e7e4b07c4_3", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"clothes drying rack\", \"references\": [{\"object\": \"clothes drying rack\", \"relation\": \"side_below\", \"distance_cm\": 7.9}, {\"object\": \"bedside table\", \"relation\": \"side_below\", \"distance_cm\": 21.0}]}", "target": "backpack", "split": "train"}
{"id": "3e7e4b07c4_4", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "3e7e4b07c4_5", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. clothes\nB. clothes drying rack\nC. blanket\nD. shirt", "choices": ["A. clothes", "B. clothes drying rack", "C. blanket", "D. shirt"], "answer": "C. blanket", "target": "bottle", "split": "train"}
{"id": "3e7e4b07c4_6", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the blanket, about 5 cm away. It is also to the side of and below the clothes drying rack, about 6 cm away. It is further to the side of and below the clothes, about 34 cm away.", "target": "bottle", "split": "train"}
{"id": "3e7e4b07c4_7", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"clothes drying rack\", \"references\": [{\"object\": \"blanket\", \"relation\": \"under\", \"distance_cm\": 4.6}, {\"object\": \"clothes drying rack\", \"relation\": \"side_below\", \"distance_cm\": 5.9}]}", "target": "bottle", "split": "train"}
{"id": "3e7e4b07c4_8", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "3e7e4b07c4_9", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the keyboard, about 9 cm away. It is further to the side of and above the headphone, about 10 cm away.", "target": "cup", "split": "train"}
{"id": "3e7e4b07c4_10", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.3}, {\"object\": \"headphone\", \"relation\": \"side_above\", \"distance_cm\": 9.8}]}", "target": "cup", "split": "train"}
{"id": "3e7e4b07c4_11", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "No", "target": "game controller", "split": "train"}
{"id": "3e7e4b07c4_12", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. bedside table\nB. backpack\nC. water pitcher\nD. bag", "choices": ["A. bedside table", "B. backpack", "C. water pitcher", "D. bag"], "answer": "C. water pitcher", "target": "extension cord", "split": "train"}
{"id": "3e7e4b07c4_13", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the water pitcher, about 24 cm away.", "target": "extension cord", "split": "train"}
{"id": "3e7e4b07c4_14", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"blender\", \"references\": [{\"object\": \"water pitcher\", \"relation\": \"side_below\", \"distance_cm\": 24.0}]}", "target": "extension cord", "split": "train"}
{"id": "3e7e4b07c4_15", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "3e7e4b07c4_16", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. cup\nB. mousepad\nC. mouse\nD. laptop holder", "choices": ["A. cup", "B. mousepad", "C. mouse", "D. laptop holder"], "answer": "C. mouse", "target": "keyboard", "split": "train"}
{"id": "3e7e4b07c4_17", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the mouse, about 5 cm away. It is further next to the laptop holder, about 7 cm away.", "target": "keyboard", "split": "train"}
{"id": "3e7e4b07c4_18", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"laptop holder\", \"relation\": \"next_to\", \"distance_cm\": 7.2}]}", "target": "keyboard", "split": "train"}
{"id": "3e7e4b07c4_19", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "3e7e4b07c4_20", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. mug\nB. keyboard\nC. laptop holder\nD. router", "choices": ["A. mug", "B. keyboard", "C. laptop holder", "D. router"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "3e7e4b07c4_21", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the keyboard, about 5 cm away. It is further next to the router, about 13 cm away.", "target": "mouse", "split": "train"}
{"id": "3e7e4b07c4_22", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"router\", \"relation\": \"next_to\", \"distance_cm\": 12.6}]}", "target": "mouse", "split": "train"}
{"id": "3e7e4b07c4_23", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "3e7e4b07c4_24", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. keyboard\nB. mug\nC. laptop holder\nD. router", "choices": ["A. keyboard", "B. mug", "C. laptop holder", "D. router"], "answer": "D. router", "target": "mousepad", "split": "train"}
{"id": "3e7e4b07c4_25", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the desk. It is also next to the router, about 7 cm away. It is further next to the keyboard, about 9 cm away.", "target": "mousepad", "split": "train"}
{"id": "3e7e4b07c4_26", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"router\", \"relation\": \"next_to\", \"distance_cm\": 6.9}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.2}]}", "target": "mousepad", "split": "train"}
{"id": "3e7e4b07c4_27", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "3e7e4b07c4_28", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the desk. It is also to the side of and below the curtain, about 17 cm away. It is further to the side of and below the cable raceway, about 17 cm away.", "target": "mug", "split": "train"}
{"id": "3e7e4b07c4_29", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 16.9}, {\"object\": \"cable raceway\", \"relation\": \"side_below\", \"distance_cm\": 17.4}]}", "target": "mug", "split": "train"}
{"id": "3e7e4b07c4_30", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "3e7e4b07c4_31", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. pillow\nB. bedside table\nC. scale\nD. shelf rail", "choices": ["A. pillow", "B. bedside table", "C. scale", "D. shelf rail"], "answer": "D. shelf rail", "target": "phone charger", "split": "train"}
{"id": "3e7e4b07c4_32", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the shelf rail, about 11 cm away. It is also to the side of and below the pillow, about 19 cm away. It is further to the side of and above the bedside table, about 23 cm away.", "target": "phone charger", "split": "train"}
{"id": "3e7e4b07c4_33", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "33", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"shelf rail\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 11.3}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 18.7}]}", "target": "phone charger", "split": "train"}
{"id": "3e7e4b07c4_34", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "34", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "3e7e4b07c4_35", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "35", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. laptop holder\nB. table lamp\nC. office chair\nD. keyboard", "choices": ["A. laptop holder", "B. table lamp", "C. office chair", "D. keyboard"], "answer": "B. table lamp", "target": "spray bottle", "split": "train"}
{"id": "3e7e4b07c4_36", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "36", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the desk. It is also to the side of and below the table lamp, about 11 cm away. It is further next to the keyboard, about 14 cm away.", "target": "spray bottle", "split": "train"}
{"id": "3e7e4b07c4_37", "scene_id": "3e7e4b07c4", "source_dataset": "scannetpp", "local_sample_id": "37", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 10.6}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 14.0}]}", "target": "spray bottle", "split": "train"}
{"id": "3e8bba0176_0", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "3e8bba0176_1", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. plate\nB. frame\nC. paper bag\nD. monitor", "choices": ["A. plate", "B. frame", "C. paper bag", "D. monitor"], "answer": "C. paper bag", "target": "bottle", "split": "train"}
{"id": "3e8bba0176_2", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the cabinet. It is also to the side of and below the paper bag, about 8 cm away. It is further near the plate, about 17 cm away.", "target": "bottle", "split": "train"}
{"id": "3e8bba0176_3", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"paper bag\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"plate\", \"relation\": \"near\", \"distance_cm\": 16.9}]}", "target": "bottle", "split": "train"}
{"id": "3e8bba0176_4", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "Yes", "target": "headphone case", "split": "train"}
{"id": "3e8bba0176_5", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphone case in this scene?\nA. table\nB. keyboard\nC. split\nD. monitor", "choices": ["A. table", "B. keyboard", "C. split", "D. monitor"], "answer": "A. table", "target": "headphone case", "split": "train"}
{"id": "3e8bba0176_6", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphone case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphone case is to the side of and above the table, about 3 cm away. It is also to the side of and above the keyboard, about 23 cm away. It is further to the side of and above the split, about 37 cm away.", "target": "headphone case", "split": "train"}
{"id": "3e8bba0176_7", "scene_id": "3e8bba0176", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphone case in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphone case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 23.1}]}", "target": "headphone case", "split": "train"}
{"id": "3e9a598891_0", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "3e9a598891_1", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. monitor\nB. monitor stand\nC. cabinet\nD. desk divider", "choices": ["A. monitor", "B. monitor stand", "C. cabinet", "D. desk divider"], "answer": "A. monitor", "target": "cup", "split": "train"}
{"id": "3e9a598891_2", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the monitor, about 6 cm away. It is also to the side of and below the desk divider, about 7 cm away. It is further to the side of and above the cabinet, about 23 cm away.", "target": "cup", "split": "train"}
{"id": "3e9a598891_3", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.9}, {\"object\": \"desk divider\", \"relation\": \"side_below\", \"distance_cm\": 7.0}]}", "target": "cup", "split": "train"}
{"id": "3e9a598891_4", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "3e9a598891_5", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. office chair\nB. monitor\nC. power socket\nD. cable duct", "choices": ["A. office chair", "B. monitor", "C. power socket", "D. cable duct"], "answer": "B. monitor", "target": "headphones", "split": "train"}
{"id": "3e9a598891_6", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and below the monitor, about 18 cm away. It is also to the side of and above the office chair, about 33 cm away. It is further to the side of and below the cable duct, about 41 cm away.", "target": "headphones", "split": "train"}
{"id": "3e9a598891_7", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.3}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 33.1}]}", "target": "headphones", "split": "train"}
{"id": "3e9a598891_8", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "3e9a598891_9", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and above the office chair, about 17 cm away. It is further near the stationaries, about 17 cm away.", "target": "keyboard", "split": "train"}
{"id": "3e9a598891_10", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 17.4}, {\"object\": \"stationaries\", \"relation\": \"near\", \"distance_cm\": 17.4}]}", "target": "keyboard", "split": "train"}
{"id": "3e9a598891_11", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "3e9a598891_12", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also to the side of and above the office chair, about 20 cm away. It is further near the box, about 21 cm away.", "target": "mouse", "split": "train"}
{"id": "3e9a598891_13", "scene_id": "3e9a598891", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 20.1}, {\"object\": \"box\", \"relation\": \"near\", \"distance_cm\": 20.7}]}", "target": "mouse", "split": "train"}
{"id": "3f6ee92fd7_0", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "3f6ee92fd7_1", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. tennis racket\nB. tissue box\nC. table\nD. printer", "choices": ["A. tennis racket", "B. tissue box", "C. table", "D. printer"], "answer": "C. table", "target": "backpack", "split": "train"}
{"id": "3f6ee92fd7_2", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the table, about 23 cm away. It is also to the side of and above the tennis racket, about 38 cm away. It is further to the side of and above the printer, about 45 cm away.", "target": "backpack", "split": "train"}
{"id": "3f6ee92fd7_3", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 23.0}, {\"object\": \"tennis racket\", \"relation\": \"side_above\", \"distance_cm\": 38.1}]}", "target": "backpack", "split": "train"}
{"id": "3f6ee92fd7_4", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "3f6ee92fd7_5", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. power strip\nB. plug\nC. monitor\nD. suitcase", "choices": ["A. power strip", "B. plug", "C. monitor", "D. suitcase"], "answer": "A. power strip", "target": "cup", "split": "train"}
{"id": "3f6ee92fd7_6", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also next to the power strip, about 6 cm away. It is further next to the plug, about 8 cm away.", "target": "cup", "split": "train"}
{"id": "3f6ee92fd7_7", "scene_id": "3f6ee92fd7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"power strip\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"plug\", \"relation\": \"next_to\", \"distance_cm\": 8.4}]}", "target": "cup", "split": "train"}
{"id": "3ff873c77e_0", "scene_id": "3ff873c77e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "3ff873c77e_1", "scene_id": "3ff873c77e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. kitchen utensils\nB. cup\nC. food\nD. plate", "choices": ["A. kitchen utensils", "B. cup", "C. food", "D. plate"], "answer": "D. plate", "target": "bottle", "split": "train"}
{"id": "3ff873c77e_2", "scene_id": "3ff873c77e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the plate, about 16 cm away. It is also to the side of and below the food, about 31 cm away. It is further to the side of and above the kitchen utensils, about 31 cm away.", "target": "bottle", "split": "train"}
{"id": "3ff873c77e_3", "scene_id": "3ff873c77e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"plate\", \"relation\": \"side_above\", \"distance_cm\": 15.9}, {\"object\": \"food\", \"relation\": \"side_below\", \"distance_cm\": 30.9}]}", "target": "bottle", "split": "train"}
{"id": "3ff873c77e_4", "scene_id": "3ff873c77e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "3ff873c77e_5", "scene_id": "3ff873c77e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and above the sink, about 6 cm away. It is also to the side of and below the plate, about 6 cm away. It is further to the side of and below the kitchen utensils, about 24 cm away.", "target": "mug", "split": "train"}
{"id": "3ff873c77e_6", "scene_id": "3ff873c77e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"plate\", \"relation\": \"side_below\", \"distance_cm\": 6.2}]}", "target": "mug", "split": "train"}
{"id": "410c470782_0", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "410c470782_1", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. keyboard\nB. chair\nC. glass case\nD. cord", "choices": ["A. keyboard", "B. chair", "C. glass case", "D. cord"], "answer": "C. glass case", "target": "bottle", "split": "train"}
{"id": "410c470782_2", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the glass case, about 5 cm away. It is further to the side of and above the cord, about 10 cm away.", "target": "bottle", "split": "train"}
{"id": "410c470782_3", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"glass case\", \"relation\": \"side_above\", \"distance_cm\": 5.1}, {\"object\": \"cord\", \"relation\": \"side_above\", \"distance_cm\": 9.6}]}", "target": "bottle", "split": "train"}
{"id": "410c470782_4", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "Yes", "target": "charger", "split": "train"}
{"id": "410c470782_5", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. door frame\nB. light switch\nC. suitcase\nD. cabinet", "choices": ["A. door frame", "B. light switch", "C. suitcase", "D. cabinet"], "answer": "B. light switch", "target": "charger", "split": "train"}
{"id": "410c470782_6", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is to the side of and below the light switch, about 30 cm away. It is also to the side of and below the door frame, about 36 cm away. It is further to the side of and above the cabinet, about 37 cm away.", "target": "charger", "split": "train"}
{"id": "410c470782_7", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 30.2}, {\"object\": \"door frame\", \"relation\": \"side_below\", \"distance_cm\": 35.5}]}", "target": "charger", "split": "train"}
{"id": "410c470782_8", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "410c470782_9", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. stuffed toy\nB. sofa\nC. mug\nD. joystick", "choices": ["A. stuffed toy", "B. sofa", "C. mug", "D. joystick"], "answer": "D. joystick", "target": "cup", "split": "train"}
{"id": "410c470782_10", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the coffee table. It is also next to the joystick, about 10 cm away. It is further near the mug, about 28 cm away.", "target": "cup", "split": "train"}
{"id": "410c470782_11", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"joystick\", \"relation\": \"next_to\", \"distance_cm\": 9.7}, {\"object\": \"mug\", \"relation\": \"near\", \"distance_cm\": 28.1}]}", "target": "cup", "split": "train"}
{"id": "410c470782_12", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "410c470782_13", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the laptop, about 8 cm away. It is further next to the laptop stand, about 9 cm away.", "target": "keyboard", "split": "train"}
{"id": "410c470782_14", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 8.4}, {\"object\": \"laptop stand\", \"relation\": \"next_to\", \"distance_cm\": 9.0}]}", "target": "keyboard", "split": "train"}
{"id": "410c470782_15", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "Yes", "target": "phone stand", "split": "train"}
{"id": "410c470782_16", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone stand in this scene?\nA. table lamp\nB. headboard pillow\nC. flower\nD. vase", "choices": ["A. table lamp", "B. headboard pillow", "C. flower", "D. vase"], "answer": "A. table lamp", "target": "phone stand", "split": "train"}
{"id": "410c470782_17", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone stand in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone stand is to the side of and below the table lamp, about 8 cm away. It is also to the side of and below the headboard pillow, about 18 cm away. It is further to the side of and below the flower, about 21 cm away.", "target": "phone stand", "split": "train"}
{"id": "410c470782_18", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone stand in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone stand\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 7.6}, {\"object\": \"headboard pillow\", \"relation\": \"side_below\", \"distance_cm\": 18.1}]}", "target": "phone stand", "split": "train"}
{"id": "410c470782_19", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "410c470782_20", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. toiletries\nB. hair tools\nC. faucet\nD. shelf", "choices": ["A. toiletries", "B. hair tools", "C. faucet", "D. shelf"], "answer": "B. hair tools", "target": "toothbrush", "split": "train"}
{"id": "410c470782_21", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is to the side of and above the hair tools, about 9 cm away. It is also near the faucet, about 16 cm away. It is further to the side of and below the shelf, about 21 cm away.", "target": "toothbrush", "split": "train"}
{"id": "410c470782_22", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"hair tools\", \"relation\": \"side_above\", \"distance_cm\": 9.3}, {\"object\": \"faucet\", \"relation\": \"near\", \"distance_cm\": 15.8}]}", "target": "toothbrush", "split": "train"}
{"id": "410c470782_23", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "410c470782_24", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. stool\nB. chair\nC. bag\nD. ukulele", "choices": ["A. stool", "B. chair", "C. bag", "D. ukulele"], "answer": "B. chair", "target": "tote bag", "split": "train"}
{"id": "410c470782_25", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is under the chair, about 4 cm away. It is also near the stool, about 27 cm away. It is further to the side of and below the ukulele, about 33 cm away.", "target": "tote bag", "split": "train"}
{"id": "410c470782_26", "scene_id": "410c470782", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"chair\", \"relation\": \"under\", \"distance_cm\": 4.0}, {\"object\": \"stool\", \"relation\": \"near\", \"distance_cm\": 26.9}]}", "target": "tote bag", "split": "train"}
{"id": "419cbe7c11_0", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "419cbe7c11_1", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bookshelf\nB. bag\nC. cabinet\nD. book", "choices": ["A. bookshelf", "B. bag", "C. cabinet", "D. book"], "answer": "C. cabinet", "target": "backpack", "split": "train"}
{"id": "419cbe7c11_2", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the cabinet, about 51 cm away.", "target": "backpack", "split": "train"}
{"id": "419cbe7c11_3", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 50.9}]}", "target": "backpack", "split": "train"}
{"id": "419cbe7c11_4", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "419cbe7c11_5", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. monitor\nB. mouse\nC. mousepad\nD. keyboard", "choices": ["A. monitor", "B. mouse", "C. mousepad", "D. keyboard"], "answer": "D. keyboard", "target": "cup", "split": "train"}
{"id": "419cbe7c11_6", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and above the keyboard, about 7 cm away. It is further to the side of and above the mousepad, about 20 cm away.", "target": "cup", "split": "train"}
{"id": "419cbe7c11_7", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 6.5}, {\"object\": \"mousepad\", \"relation\": \"side_above\", \"distance_cm\": 20.3}]}", "target": "cup", "split": "train"}
{"id": "419cbe7c11_8", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "419cbe7c11_9", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the outlet, about 5 cm away. It is further next to the power socket, about 6 cm away.", "target": "keyboard", "split": "train"}
{"id": "419cbe7c11_10", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"outlet\", \"relation\": \"side_below\", \"distance_cm\": 5.1}, {\"object\": \"power socket\", \"relation\": \"next_to\", \"distance_cm\": 6.0}]}", "target": "keyboard", "split": "train"}
{"id": "419cbe7c11_11", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "419cbe7c11_12", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. outlet\nB. table\nC. windowsill\nD. tissue box", "choices": ["A. outlet", "B. table", "C. windowsill", "D. tissue box"], "answer": "A. outlet", "target": "mouse", "split": "train"}
{"id": "419cbe7c11_13", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the outlet, about 3 cm away. It is further next to the tissue box, about 8 cm away.", "target": "mouse", "split": "train"}
{"id": "419cbe7c11_14", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"outlet\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"tissue box\", \"relation\": \"next_to\", \"distance_cm\": 7.6}]}", "target": "mouse", "split": "train"}
{"id": "419cbe7c11_15", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "419cbe7c11_16", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. monitor\nB. cabinet\nC. rolling chair\nD. keyboard", "choices": ["A. monitor", "B. cabinet", "C. rolling chair", "D. keyboard"], "answer": "C. rolling chair", "target": "mousepad", "split": "train"}
{"id": "419cbe7c11_17", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is to the side of and above the rolling chair, about 6 cm away. It is also next to the keyboard, about 8 cm away. It is further to the side of and above the cabinet, about 10 cm away.", "target": "mousepad", "split": "train"}
{"id": "419cbe7c11_18", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"rolling chair\", \"relation\": \"side_above\", \"distance_cm\": 6.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.5}]}", "target": "mousepad", "split": "train"}
{"id": "419cbe7c11_19", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "419cbe7c11_20", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. bag\nB. book\nC. bookshelf\nD. backpack", "choices": ["A. bag", "B. book", "C. bookshelf", "D. backpack"], "answer": "A. bag", "target": "mug", "split": "train"}
{"id": "419cbe7c11_21", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and above the bag, about 42 cm away.", "target": "mug", "split": "train"}
{"id": "419cbe7c11_22", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 41.5}]}", "target": "mug", "split": "train"}
{"id": "419cbe7c11_23", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "419cbe7c11_24", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. table\nB. monitor\nC. paper\nD. tube", "choices": ["A. table", "B. monitor", "C. paper", "D. tube"], "answer": "A. table", "target": "pen", "split": "train"}
{"id": "419cbe7c11_25", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the table. It is also to the side of and below the monitor, about 12 cm away. It is further next to the tube, about 14 cm away.", "target": "pen", "split": "train"}
{"id": "419cbe7c11_26", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.2}, {\"object\": \"tube\", \"relation\": \"next_to\", \"distance_cm\": 14.0}]}", "target": "pen", "split": "train"}
{"id": "419cbe7c11_27", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "419cbe7c11_28", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. mug\nB. book\nC. mouse\nD. table", "choices": ["A. mug", "B. book", "C. mouse", "D. table"], "answer": "A. mug", "target": "tissue box", "split": "train"}
{"id": "419cbe7c11_29", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also next to the mug, about 6 cm away. It is further next to the mouse, about 8 cm away.", "target": "tissue box", "split": "train"}
{"id": "419cbe7c11_30", "scene_id": "419cbe7c11", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 5.8}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 7.6}]}", "target": "tissue box", "split": "train"}
{"id": "41eb967018_0", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "41eb967018_1", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. keyboard\nC. table\nD. telephone", "choices": ["A. monitor", "B. keyboard", "C. table", "D. telephone"], "answer": "D. telephone", "target": "bottle", "split": "train"}
{"id": "41eb967018_2", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the telephone, about 15 cm away. It is further to the side of and below the monitor, about 23 cm away.", "target": "bottle", "split": "train"}
{"id": "41eb967018_3", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"telephone\", \"relation\": \"side_above\", \"distance_cm\": 14.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.3}]}", "target": "bottle", "split": "train"}
{"id": "41eb967018_4", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "No", "target": "game controller", "split": "train"}
{"id": "41eb967018_5", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. office chair\nC. computer tower\nD. mouse", "choices": ["A. monitor", "B. office chair", "C. computer tower", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "train"}
{"id": "41eb967018_6", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse. It is also to the side of and below the monitor, about 8 cm away. It is further to the side of and above the office chair, about 24 cm away.", "target": "keyboard", "split": "train"}
{"id": "41eb967018_7", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.1}]}", "target": "keyboard", "split": "train"}
{"id": "41eb967018_8", "scene_id": "41eb967018", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "4291be3b44_0", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "4291be3b44_1", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. mirror\nB. socket\nC. blanket\nD. electric heater", "choices": ["A. mirror", "B. socket", "C. blanket", "D. electric heater"], "answer": "A. mirror", "target": "backpack", "split": "train"}
{"id": "4291be3b44_2", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the rug. It is also to the side of and below the mirror, about 26 cm away. It is further to the side of and above the socket, about 43 cm away.", "target": "backpack", "split": "train"}
{"id": "4291be3b44_3", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"rug\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 25.8}, {\"object\": \"socket\", \"relation\": \"side_above\", \"distance_cm\": 43.3}]}", "target": "backpack", "split": "train"}
{"id": "4291be3b44_4", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "4291be3b44_5", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. storage bin\nB. wall lamp\nC. book\nD. shelf", "choices": ["A. storage bin", "B. wall lamp", "C. book", "D. shelf"], "answer": "C. book", "target": "bottle", "split": "train"}
{"id": "4291be3b44_6", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the shelf, about 20 cm away. It is also to the side of and above the storage bin, about 30 cm away. It is further to the side of and below the wall lamp, about 57 cm away.", "target": "bottle", "split": "train"}
{"id": "4291be3b44_7", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 20.5}, {\"object\": \"storage bin\", \"relation\": \"side_above\", \"distance_cm\": 29.8}]}", "target": "bottle", "split": "train"}
{"id": "4291be3b44_8", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "4291be3b44_9", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. toiletries\nB. papers\nC. blanket\nD. storage bin", "choices": ["A. toiletries", "B. papers", "C. blanket", "D. storage bin"], "answer": "D. storage bin", "target": "cup", "split": "train"}
{"id": "4291be3b44_10", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the storage bin, about 3 cm away. It is also next to the toiletries, about 6 cm away. It is further to the side of and above the blanket, about 26 cm away.", "target": "cup", "split": "train"}
{"id": "4291be3b44_11", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"storage bin\", \"relation\": \"next_to\", \"distance_cm\": 2.8}, {\"object\": \"toiletries\", \"relation\": \"next_to\", \"distance_cm\": 5.8}]}", "target": "cup", "split": "train"}
{"id": "4291be3b44_12", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "4291be3b44_13", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. papers\nB. paper\nC. mouse\nD. books", "choices": ["A. papers", "B. paper", "C. mouse", "D. books"], "answer": "C. mouse", "target": "keyboard", "split": "train"}
{"id": "4291be3b44_14", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse, about 3 cm away. It is also next to the papers, about 8 cm away. It is further to the side of and below the paper, about 15 cm away.", "target": "keyboard", "split": "train"}
{"id": "4291be3b44_15", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 8.1}]}", "target": "keyboard", "split": "train"}
{"id": "4291be3b44_16", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "4291be3b44_17", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. paper\nB. books\nC. papers\nD. keyboard", "choices": ["A. paper", "B. books", "C. papers", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "4291be3b44_18", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 3 cm away. It is also next to the papers, about 5 cm away. It is further under the books, about 5 cm away.", "target": "mouse", "split": "train"}
{"id": "4291be3b44_19", "scene_id": "4291be3b44", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 4.5}]}", "target": "mouse", "split": "train"}
{"id": "4318f8bb3c_0", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "4318f8bb3c_1", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. chair\nB. cloth\nC. cutting board\nD. suitcase", "choices": ["A. chair", "B. cloth", "C. cutting board", "D. suitcase"], "answer": "B. cloth", "target": "bottle", "split": "train"}
{"id": "4318f8bb3c_2", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the cloth, about 9 cm away. It is further to the side of and above the suitcase, about 14 cm away.", "target": "bottle", "split": "train"}
{"id": "4318f8bb3c_3", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cloth\", \"relation\": \"side_above\", \"distance_cm\": 8.6}, {\"object\": \"suitcase\", \"relation\": \"side_above\", \"distance_cm\": 13.7}]}", "target": "bottle", "split": "train"}
{"id": "4318f8bb3c_4", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "4318f8bb3c_5", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. laptop\nB. cabinet\nC. monitor\nD. chair", "choices": ["A. laptop", "B. cabinet", "C. monitor", "D. chair"], "answer": "A. laptop", "target": "keyboard", "split": "train"}
{"id": "4318f8bb3c_6", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is under the laptop, about 3 cm away. It is also to the side of and below the monitor, about 6 cm away. It is further to the side of and above the cabinet, about 19 cm away.", "target": "keyboard", "split": "train"}
{"id": "4318f8bb3c_7", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"tabletop\", \"references\": [{\"object\": \"laptop\", \"relation\": \"under\", \"distance_cm\": 3.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.0}]}", "target": "keyboard", "split": "train"}
{"id": "4318f8bb3c_8", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "4318f8bb3c_9", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. cabinet\nB. keyboard\nC. monitor\nD. book", "choices": ["A. cabinet", "B. keyboard", "C. monitor", "D. book"], "answer": "C. monitor", "target": "mug", "split": "train"}
{"id": "4318f8bb3c_10", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and below the monitor, about 12 cm away. It is also to the side of and above the keyboard, about 26 cm away. It is further near the book, about 28 cm away.", "target": "mug", "split": "train"}
{"id": "4318f8bb3c_11", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"tabletop\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.4}, {\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 26.3}]}", "target": "mug", "split": "train"}
{"id": "4318f8bb3c_12", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "4318f8bb3c_13", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. rice cooker\nB. chair\nC. cutting board\nD. water filter", "choices": ["A. rice cooker", "B. chair", "C. cutting board", "D. water filter"], "answer": "C. cutting board", "target": "water bottle", "split": "train"}
{"id": "4318f8bb3c_14", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also to the side of and above the cutting board, about 13 cm away. It is further to the side of and above the chair, about 15 cm away.", "target": "water bottle", "split": "train"}
{"id": "4318f8bb3c_15", "scene_id": "4318f8bb3c", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cutting board\", \"relation\": \"side_above\", \"distance_cm\": 13.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 15.1}]}", "target": "water bottle", "split": "train"}
{"id": "4380e4646a_0", "scene_id": "4380e4646a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "4380e4646a_1", "scene_id": "4380e4646a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the stove top. It is also under the cutting boards, about 3 cm away. It is further to the side of and above the frying pan, about 12 cm away.", "target": "bottle", "split": "train"}
{"id": "4380e4646a_2", "scene_id": "4380e4646a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"stove top\", \"relation\": \"side_above\", \"distance_cm\": 2.4}, {\"object\": \"cutting boards\", \"relation\": \"under\", \"distance_cm\": 2.5}]}", "target": "bottle", "split": "train"}
{"id": "4380e4646a_3", "scene_id": "4380e4646a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "4380e4646a_4", "scene_id": "4380e4646a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. mat\nB. sandles\nC. shoe rack\nD. door frame", "choices": ["A. mat", "B. sandles", "C. shoe rack", "D. door frame"], "answer": "D. door frame", "target": "power adapter", "split": "train"}
{"id": "4380e4646a_5", "scene_id": "4380e4646a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and below the door frame, about 11 cm away. It is also to the side of and above the mat, about 28 cm away. It is further to the side of and above the shoe rack, about 42 cm away.", "target": "power adapter", "split": "train"}
{"id": "4380e4646a_6", "scene_id": "4380e4646a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"shoe rack\", \"references\": [{\"object\": \"door frame\", \"relation\": \"side_below\", \"distance_cm\": 11.2}, {\"object\": \"mat\", \"relation\": \"side_above\", \"distance_cm\": 27.9}]}", "target": "power adapter", "split": "train"}
{"id": "44c85584ae_0", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "44c85584ae_1", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothes\nB. mirror\nC. door frame\nD. jacket", "choices": ["A. clothes", "B. mirror", "C. door frame", "D. jacket"], "answer": "C. door frame", "target": "backpack", "split": "train"}
{"id": "44c85584ae_2", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is attached to the door. It is also to the side of and below the door frame, about 10 cm away. It is further to the side of and above the mirror, about 54 cm away.", "target": "backpack", "split": "train"}
{"id": "44c85584ae_3", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door frame\", \"relation\": \"side_below\", \"distance_cm\": 9.5}, {\"object\": \"mirror\", \"relation\": \"side_above\", \"distance_cm\": 53.8}]}", "target": "backpack", "split": "train"}
{"id": "44c85584ae_4", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "44c85584ae_5", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. bedside drawer unit\nB. clothes\nC. backpack\nD. pillow", "choices": ["A. bedside drawer unit", "B. clothes", "C. backpack", "D. pillow"], "answer": "D. pillow", "target": "extension cord", "split": "train"}
{"id": "44c85584ae_6", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the pillow, about 49 cm away.", "target": "extension cord", "split": "train"}
{"id": "44c85584ae_7", "scene_id": "44c85584ae", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"bedside drawer unit\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 48.6}]}", "target": "extension cord", "split": "train"}
{"id": "4517d988d8_0", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "4517d988d8_1", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chandelier\nB. audio amplifier\nC. bottle\nD. chair", "choices": ["A. chandelier", "B. audio amplifier", "C. bottle", "D. chair"], "answer": "B. audio amplifier", "target": "backpack", "split": "train"}
{"id": "4517d988d8_2", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the audio amplifier, about 47 cm away.", "target": "backpack", "split": "train"}
{"id": "4517d988d8_3", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"audio amplifier\", \"relation\": \"side_below\", \"distance_cm\": 47.1}]}", "target": "backpack", "split": "train"}
{"id": "4517d988d8_4", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "4517d988d8_5", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. plate\nB. chandelier\nC. table\nD. chair", "choices": ["A. plate", "B. chandelier", "C. table", "D. chair"], "answer": "C. table", "target": "bottle", "split": "train"}
{"id": "4517d988d8_6", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the table, about 8 cm away. It is also to the side of and above the plate, about 12 cm away. It is further to the side of and below the chandelier, about 34 cm away.", "target": "bottle", "split": "train"}
{"id": "4517d988d8_7", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 8.3}, {\"object\": \"plate\", \"relation\": \"side_above\", \"distance_cm\": 12.5}]}", "target": "bottle", "split": "train"}
{"id": "4517d988d8_8", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "4517d988d8_9", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. light switches\nB. audio amplifier\nC. bottle\nD. backpack", "choices": ["A. light switches", "B. audio amplifier", "C. bottle", "D. backpack"], "answer": "A. light switches", "target": "phone charger", "split": "train"}
{"id": "4517d988d8_10", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is below the light switches, about 74 cm away.", "target": "phone charger", "split": "train"}
{"id": "4517d988d8_11", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"power socket\", \"references\": [{\"object\": \"light switches\", \"relation\": \"below\", \"distance_cm\": 73.6}]}", "target": "phone charger", "split": "train"}
{"id": "4517d988d8_12", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "train"}
{"id": "4517d988d8_13", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the dresser.", "target": "tv remote", "split": "train"}
{"id": "4517d988d8_14", "scene_id": "4517d988d8", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"dresser\", \"references\": []}", "target": "tv remote", "split": "train"}
{"id": "45b0dac5e3_0", "scene_id": "45b0dac5e3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "45b0dac5e3_1", "scene_id": "45b0dac5e3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. toothpaste\nB. towel\nC. mirror\nD. bottle", "choices": ["A. toothpaste", "B. towel", "C. mirror", "D. bottle"], "answer": "D. bottle", "target": "cup", "split": "train"}
{"id": "45b0dac5e3_2", "scene_id": "45b0dac5e3", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the bottle, about 14 cm away. It is also below the mirror, about 32 cm away. It is further near the toothpaste, about 33 cm away.", "target": "cup", "split": "train"}
{"id": "45b0dac5e3_3", "scene_id": "45b0dac5e3", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"bathroom cabinet\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 13.7}, {\"object\": \"mirror\", \"relation\": \"below\", \"distance_cm\": 32.1}]}", "target": "cup", "split": "train"}
{"id": "45ba9ea301_0", "scene_id": "45ba9ea301", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "45ba9ea301_1", "scene_id": "45ba9ea301", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. tupperware\nB. kitchen towel\nC. paper towel roll\nD. microwave", "choices": ["A. tupperware", "B. kitchen towel", "C. paper towel roll", "D. microwave"], "answer": "A. tupperware", "target": "cup", "split": "train"}
{"id": "45ba9ea301_2", "scene_id": "45ba9ea301", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the tupperware, about 4 cm away. It is also to the side of and below the microwave, about 18 cm away. It is further to the side of and below the paper towel roll, about 21 cm away.", "target": "cup", "split": "train"}
{"id": "45ba9ea301_3", "scene_id": "45ba9ea301", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"tupperware\", \"relation\": \"next_to\", \"distance_cm\": 4.4}, {\"object\": \"microwave\", \"relation\": \"side_below\", \"distance_cm\": 18.2}]}", "target": "cup", "split": "train"}
{"id": "4610b2104c_0", "scene_id": "4610b2104c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "4610b2104c_1", "scene_id": "4610b2104c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. lamp\nB. paper towel\nC. office chair\nD. power sockets", "choices": ["A. lamp", "B. paper towel", "C. office chair", "D. power sockets"], "answer": "C. office chair", "target": "power strip", "split": "train"}
{"id": "4610b2104c_2", "scene_id": "4610b2104c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is near the office chair, about 34 cm away. It is also to the side of and below the lamp, about 38 cm away. It is further to the side of and below the paper towel, about 60 cm away.", "target": "power strip", "split": "train"}
{"id": "4610b2104c_3", "scene_id": "4610b2104c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"near\", \"distance_cm\": 33.7}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 37.7}]}", "target": "power strip", "split": "train"}
{"id": "46638cfd0f_0", "scene_id": "46638cfd0f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "46638cfd0f_1", "scene_id": "46638cfd0f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil in this scene?\nA. towel\nB. standing lamp\nC. kitchen cabinet\nD. egg carton", "choices": ["A. towel", "B. standing lamp", "C. kitchen cabinet", "D. egg carton"], "answer": "B. standing lamp", "target": "pencil", "split": "train"}
{"id": "46638cfd0f_2", "scene_id": "46638cfd0f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil is to the side of and above the standing lamp, about 28 cm away. It is also near the egg carton, about 29 cm away. It is further to the side of and above the kitchen cabinet, about 41 cm away.", "target": "pencil", "split": "train"}
{"id": "46638cfd0f_3", "scene_id": "46638cfd0f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"standing lamp\", \"relation\": \"side_above\", \"distance_cm\": 27.9}, {\"object\": \"egg carton\", \"relation\": \"near\", \"distance_cm\": 29.1}]}", "target": "pencil", "split": "train"}
{"id": "480ddaadc0_0", "scene_id": "480ddaadc0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "480ddaadc0_1", "scene_id": "480ddaadc0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone in this scene?\nA. monitor\nB. cabinet\nC. radiator\nD. window sill", "choices": ["A. monitor", "B. cabinet", "C. radiator", "D. window sill"], "answer": "A. monitor", "target": "phone", "split": "train"}
{"id": "480ddaadc0_2", "scene_id": "480ddaadc0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone is on the desk. It is also to the side of and below the monitor, about 9 cm away. It is further to the side of and below the window sill, about 11 cm away.", "target": "phone", "split": "train"}
{"id": "480ddaadc0_3", "scene_id": "480ddaadc0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.4}, {\"object\": \"window sill\", \"relation\": \"side_below\", \"distance_cm\": 10.8}]}", "target": "phone", "split": "train"}
{"id": "484ad681df_0", "scene_id": "484ad681df", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "484ad681df_1", "scene_id": "484ad681df", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. shoe\nC. bed\nD. projector", "choices": ["A. monitor", "B. shoe", "C. bed", "D. projector"], "answer": "D. projector", "target": "backpack", "split": "train"}
{"id": "484ad681df_2", "scene_id": "484ad681df", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the projector, about 4 cm away. It is also near the shoe, about 67 cm away.", "target": "backpack", "split": "train"}
{"id": "484ad681df_3", "scene_id": "484ad681df", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"projector\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"shoe\", \"relation\": \"near\", \"distance_cm\": 67.4}]}", "target": "backpack", "split": "train"}
{"id": "4897e95232_0", "scene_id": "4897e95232", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "4897e95232_1", "scene_id": "4897e95232", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. trash can\nB. landline phone\nC. cabinet\nD. curtain", "choices": ["A. trash can", "B. landline phone", "C. cabinet", "D. curtain"], "answer": "D. curtain", "target": "bottle", "split": "train"}
{"id": "4897e95232_2", "scene_id": "4897e95232", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the cabinet. It is also to the side of and below the curtain, about 5 cm away. It is further to the side of and above the trash can, about 57 cm away.", "target": "bottle", "split": "train"}
{"id": "4897e95232_3", "scene_id": "4897e95232", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 5.1}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 57.5}]}", "target": "bottle", "split": "train"}
{"id": "49fe768553_0", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "49fe768553_1", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. pillow\nB. table\nC. bottle\nD. yoga mat", "choices": ["A. pillow", "B. table", "C. bottle", "D. yoga mat"], "answer": "C. bottle", "target": "backpack", "split": "train"}
{"id": "49fe768553_2", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the bottle, about 3 cm away. It is also to the side of and below the pillow, about 10 cm away. It is further to the side of and above the yoga mat, about 22 cm away.", "target": "backpack", "split": "train"}
{"id": "49fe768553_3", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 2.6}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 10.2}]}", "target": "backpack", "split": "train"}
{"id": "49fe768553_4", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "49fe768553_5", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. guitar\nB. remote\nC. pedestal fan\nD. book", "choices": ["A. guitar", "B. remote", "C. pedestal fan", "D. book"], "answer": "C. pedestal fan", "target": "bottle", "split": "train"}
{"id": "49fe768553_6", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the pedestal fan, about 8 cm away. It is also to the side of and above the book, about 13 cm away. It is further to the side of and below the remote, about 16 cm away.", "target": "bottle", "split": "train"}
{"id": "49fe768553_7", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"side table\", \"references\": [{\"object\": \"pedestal fan\", \"relation\": \"side_below\", \"distance_cm\": 8.2}, {\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 13.2}]}", "target": "bottle", "split": "train"}
{"id": "49fe768553_8", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "Yes", "target": "glasses case", "split": "train"}
{"id": "49fe768553_9", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. remote\nB. tv\nC. guitar\nD. book", "choices": ["A. remote", "B. tv", "C. guitar", "D. book"], "answer": "A. remote", "target": "glasses case", "split": "train"}
{"id": "49fe768553_10", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is next to the remote, about 5 cm away. It is also to the side of and below the tv, about 21 cm away. It is further to the side of and below the guitar, about 22 cm away.", "target": "glasses case", "split": "train"}
{"id": "49fe768553_11", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"side table\", \"references\": [{\"object\": \"remote\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 20.6}]}", "target": "glasses case", "split": "train"}
{"id": "49fe768553_12", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "49fe768553_13", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the speaker, about 4 cm away. It is also next to the wallet, about 7 cm away. It is further to the side of and above the office chair, about 10 cm away.", "target": "headphones", "split": "train"}
{"id": "49fe768553_14", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk lamp\", \"references\": [{\"object\": \"speaker\", \"relation\": \"next_to\", \"distance_cm\": 4.2}, {\"object\": \"wallet\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "headphones", "split": "train"}
{"id": "49fe768553_15", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "No", "target": "soap", "split": "train"}
{"id": "49fe768553_16", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. book\nB. office chair\nC. headphones\nD. wallet", "choices": ["A. book", "B. office chair", "C. headphones", "D. wallet"], "answer": "B. office chair", "target": "keyboard", "split": "train"}
{"id": "49fe768553_17", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the office chair, about 4 cm away. It is also next to the book, about 7 cm away. It is further next to the wallet, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "49fe768553_18", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 7.3}]}", "target": "keyboard", "split": "train"}
{"id": "49fe768553_19", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "49fe768553_20", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the cabinet. It is also next to the decoration, about 10 cm away. It is further to the side of and above the food packet, about 10 cm away.", "target": "mug", "split": "train"}
{"id": "49fe768553_21", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"decoration\", \"relation\": \"next_to\", \"distance_cm\": 10.0}, {\"object\": \"food packet\", \"relation\": \"side_above\", \"distance_cm\": 10.2}]}", "target": "mug", "split": "train"}
{"id": "49fe768553_22", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "49fe768553_23", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. rice cooker\nB. power strip\nC. backpack\nD. heater", "choices": ["A. rice cooker", "B. power strip", "C. backpack", "D. heater"], "answer": "D. heater", "target": "power adapter", "split": "train"}
{"id": "49fe768553_24", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and below the heater, about 15 cm away. It is also near the power strip, about 45 cm away. It is further to the side of and below the rice cooker, about 51 cm away.", "target": "power adapter", "split": "train"}
{"id": "49fe768553_25", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"heater\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 15.3}, {\"object\": \"power strip\", \"relation\": \"near\", \"distance_cm\": 44.8}]}", "target": "power adapter", "split": "train"}
{"id": "49fe768553_26", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "49fe768553_27", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. clothing iron\nB. side table\nC. cabinet\nD. book", "choices": ["A. clothing iron", "B. side table", "C. cabinet", "D. book"], "answer": "B. side table", "target": "power strip", "split": "train"}
{"id": "49fe768553_28", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the side table, about 8 cm away. It is also to the side of and below the book, about 10 cm away. It is further to the side of and below the cabinet, about 15 cm away.", "target": "power strip", "split": "train"}
{"id": "49fe768553_29", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"side table\", \"references\": [{\"object\": \"side table\", \"relation\": \"side_below\", \"distance_cm\": 8.1}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 10.2}]}", "target": "power strip", "split": "train"}
{"id": "49fe768553_30", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "Yes", "target": "remote", "split": "train"}
{"id": "49fe768553_31", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. glasses case\nB. bottle\nC. book\nD. tv", "choices": ["A. glasses case", "B. bottle", "C. book", "D. tv"], "answer": "A. glasses case", "target": "remote", "split": "train"}
{"id": "49fe768553_32", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is next to the glasses case, about 5 cm away. It is also to the side of and above the bottle, about 16 cm away. It is further to the side of and below the tv, about 22 cm away.", "target": "remote", "split": "train"}
{"id": "49fe768553_33", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "33", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"side table\", \"references\": [{\"object\": \"glasses case\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 16.2}]}", "target": "remote", "split": "train"}
{"id": "49fe768553_34", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "34", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "49fe768553_35", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "35", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the wallet in this scene?\nA. book\nB. keyboard\nC. speaker\nD. headphones", "choices": ["A. book", "B. keyboard", "C. speaker", "D. headphones"], "answer": "A. book", "target": "wallet", "split": "train"}
{"id": "49fe768553_36", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "36", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the wallet in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The wallet is next to the headphones, about 7 cm away. It is also next to the keyboard, about 10 cm away. It is further next to the speaker, about 12 cm away.", "target": "wallet", "split": "train"}
{"id": "49fe768553_37", "scene_id": "49fe768553", "source_dataset": "scannetpp", "local_sample_id": "37", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the wallet in structured JSON format.", "choices": null, "answer": "{\"target\": \"wallet\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 7.1}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 10.1}]}", "target": "wallet", "split": "train"}
{"id": "4a1a3a7dc5_0", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "4a1a3a7dc5_1", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. cabinet\nB. bag\nC. box\nD. shoes", "choices": ["A. cabinet", "B. bag", "C. box", "D. shoes"], "answer": "D. shoes", "target": "bottle", "split": "train"}
{"id": "4a1a3a7dc5_2", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the shoes, about 75 cm away.", "target": "bottle", "split": "train"}
{"id": "4a1a3a7dc5_3", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"eraser\", \"references\": [{\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 74.6}]}", "target": "bottle", "split": "train"}
{"id": "4a1a3a7dc5_4", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "4a1a3a7dc5_5", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. tea pot\nB. monitor\nC. keyboard\nD. papers", "choices": ["A. tea pot", "B. monitor", "C. keyboard", "D. papers"], "answer": "B. monitor", "target": "cup", "split": "train"}
{"id": "4a1a3a7dc5_6", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and below the monitor, about 7 cm away. It is further next to the tea pot, about 11 cm away.", "target": "cup", "split": "train"}
{"id": "4a1a3a7dc5_7", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.7}, {\"object\": \"tea pot\", \"relation\": \"next_to\", \"distance_cm\": 10.7}]}", "target": "cup", "split": "train"}
{"id": "4a1a3a7dc5_8", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "4a1a3a7dc5_9", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the eraser in this scene?\nA. bottle\nB. bag\nC. box\nD. shoes", "choices": ["A. bottle", "B. bag", "C. box", "D. shoes"], "answer": "D. shoes", "target": "eraser", "split": "train"}
{"id": "4a1a3a7dc5_10", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the eraser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The eraser is to the side of and above the shoes, about 81 cm away.", "target": "eraser", "split": "train"}
{"id": "4a1a3a7dc5_11", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the eraser in structured JSON format.", "choices": null, "answer": "{\"target\": \"eraser\", \"support_surface\": \"bottle\", \"references\": [{\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 81.2}]}", "target": "eraser", "split": "train"}
{"id": "4a1a3a7dc5_12", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "4a1a3a7dc5_13", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. cabinet\nB. tissues\nC. pen\nD. paper", "choices": ["A. cabinet", "B. tissues", "C. pen", "D. paper"], "answer": "D. paper", "target": "glasses case", "split": "train"}
{"id": "4a1a3a7dc5_14", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is on the table. It is also next to the pen, about 11 cm away. It is further near the tissues, about 18 cm away.", "target": "glasses case", "split": "train"}
{"id": "4a1a3a7dc5_15", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pen\", \"relation\": \"next_to\", \"distance_cm\": 10.8}, {\"object\": \"tissues\", \"relation\": \"near\", \"distance_cm\": 17.6}]}", "target": "glasses case", "split": "train"}
{"id": "4a1a3a7dc5_16", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "Yes", "target": "highlighter", "split": "train"}
{"id": "4a1a3a7dc5_17", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the highlighter in this scene?\nA. keyboard\nB. table\nC. pen\nD. box", "choices": ["A. keyboard", "B. table", "C. pen", "D. box"], "answer": "D. box", "target": "highlighter", "split": "train"}
{"id": "4a1a3a7dc5_18", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the highlighter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The highlighter is next to the box, about 6 cm away. It is also near the keyboard, about 15 cm away. It is further near the pen, about 33 cm away.", "target": "highlighter", "split": "train"}
{"id": "4a1a3a7dc5_19", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the highlighter in structured JSON format.", "choices": null, "answer": "{\"target\": \"highlighter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.5}]}", "target": "highlighter", "split": "train"}
{"id": "4a1a3a7dc5_20", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "4a1a3a7dc5_21", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. chair\nB. monitor\nC. mouse\nD. table", "choices": ["A. chair", "B. monitor", "C. mouse", "D. table"], "answer": "A. chair", "target": "notebook", "split": "train"}
{"id": "4a1a3a7dc5_22", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also to the side of and above the chair, about 14 cm away. It is further near the mouse, about 29 cm away.", "target": "notebook", "split": "train"}
{"id": "4a1a3a7dc5_23", "scene_id": "4a1a3a7dc5", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 14.2}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 28.5}]}", "target": "notebook", "split": "train"}
{"id": "4aef651da7_0", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lanyard appear?", "choices": null, "answer": "No", "target": "lanyard", "split": "train"}
{"id": "4aef651da7_1", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. office chair\nB. laptop\nC. power socket\nD. desk", "choices": ["A. office chair", "B. laptop", "C. power socket", "D. desk"], "answer": "D. desk", "target": "backpack", "split": "train"}
{"id": "4aef651da7_2", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the desk. It is also to the side of and above the office chair, about 9 cm away. It is further to the side of and above the laptop, about 24 cm away.", "target": "backpack", "split": "train"}
{"id": "4aef651da7_3", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 9.2}, {\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 24.0}]}", "target": "backpack", "split": "train"}
{"id": "4aef651da7_4", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "4aef651da7_5", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. keyboard\nB. conduit\nC. laptop\nD. book", "choices": ["A. keyboard", "B. conduit", "C. laptop", "D. book"], "answer": "D. book", "target": "headphones", "split": "train"}
{"id": "4aef651da7_6", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the desk. It is also next to the book, about 3 cm away. It is further next to the laptop, about 10 cm away.", "target": "headphones", "split": "train"}
{"id": "4aef651da7_7", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 10.3}]}", "target": "headphones", "split": "train"}
{"id": "4aef651da7_8", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "4aef651da7_9", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. laptop\nB. headphones\nC. book\nD. mouse", "choices": ["A. laptop", "B. headphones", "C. book", "D. mouse"], "answer": "A. laptop", "target": "keyboard", "split": "train"}
{"id": "4aef651da7_10", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the laptop. It is further next to the mouse, about 5 cm away.", "target": "keyboard", "split": "train"}
{"id": "4aef651da7_11", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.3}]}", "target": "keyboard", "split": "train"}
{"id": "4aef651da7_12", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "4aef651da7_13", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. laptop\nB. chair\nC. keyboard\nD. plastic bottle", "choices": ["A. laptop", "B. chair", "C. keyboard", "D. plastic bottle"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "4aef651da7_14", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the keyboard, about 5 cm away. It is further next to the laptop, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "4aef651da7_15", "scene_id": "4aef651da7", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.3}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 8.6}]}", "target": "mouse", "split": "train"}
{"id": "4ba22fa7e4_0", "scene_id": "4ba22fa7e4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "train"}
{"id": "4ba22fa7e4_1", "scene_id": "4ba22fa7e4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and below the monitor, about 29 cm away. It is further to the side of and above the keyboard, about 29 cm away.", "target": "bottle", "split": "train"}
{"id": "4ba22fa7e4_2", "scene_id": "4ba22fa7e4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 28.7}, {\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 28.9}]}", "target": "bottle", "split": "train"}
{"id": "4ba22fa7e4_3", "scene_id": "4ba22fa7e4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "4ba22fa7e4_4", "scene_id": "4ba22fa7e4", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. split\nB. bottle\nC. telephone\nD. monitor", "choices": ["A. split", "B. bottle", "C. telephone", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "4ba22fa7e4_5", "scene_id": "4ba22fa7e4", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the monitor, about 24 cm away. It is further to the side of and below the bottle, about 29 cm away.", "target": "keyboard", "split": "train"}
{"id": "4ba22fa7e4_6", "scene_id": "4ba22fa7e4", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 24.4}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 28.9}]}", "target": "keyboard", "split": "train"}
{"id": "4c141d5b1b_0", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "4c141d5b1b_1", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. flower\nB. plant pot\nC. heater\nD. cup", "choices": ["A. flower", "B. plant pot", "C. heater", "D. cup"], "answer": "D. cup", "target": "bottle", "split": "train"}
{"id": "4c141d5b1b_2", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the cup, about 4 cm away. It is also next to the flower, about 11 cm away. It is further to the side of and above the plant pot, about 13 cm away.", "target": "bottle", "split": "train"}
{"id": "4c141d5b1b_3", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"window frame\", \"references\": [{\"object\": \"cup\", \"relation\": \"side_above\", \"distance_cm\": 3.9}, {\"object\": \"flower\", \"relation\": \"next_to\", \"distance_cm\": 11.1}]}", "target": "bottle", "split": "train"}
{"id": "4c141d5b1b_4", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "4c141d5b1b_5", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. chair\nB. bottle\nC. heat pipe\nD. notebook", "choices": ["A. chair", "B. bottle", "C. heat pipe", "D. notebook"], "answer": "A. chair", "target": "extension cord", "split": "train"}
{"id": "4c141d5b1b_6", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is under the chair, about 4 cm away. It is also to the side of and below the heat pipe, about 23 cm away. It is further to the side of and below the bottle, about 69 cm away.", "target": "extension cord", "split": "train"}
{"id": "4c141d5b1b_7", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"under\", \"distance_cm\": 3.6}, {\"object\": \"heat pipe\", \"relation\": \"side_below\", \"distance_cm\": 23.0}]}", "target": "extension cord", "split": "train"}
{"id": "4c141d5b1b_8", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lanyard appear?", "choices": null, "answer": "No", "target": "lanyard", "split": "train"}
{"id": "4c141d5b1b_9", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. vacuum cleaner\nB. papers\nC. cup\nD. wardrobe", "choices": ["A. vacuum cleaner", "B. papers", "C. cup", "D. wardrobe"], "answer": "B. papers", "target": "headphones", "split": "train"}
{"id": "4c141d5b1b_10", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the bookshelf. It is also next to the papers, about 5 cm away. It is further to the side of and above the vacuum cleaner, about 13 cm away.", "target": "headphones", "split": "train"}
{"id": "4c141d5b1b_11", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 4.9}, {\"object\": \"vacuum cleaner\", \"relation\": \"side_above\", \"distance_cm\": 12.5}]}", "target": "headphones", "split": "train"}
{"id": "4c141d5b1b_12", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "4c141d5b1b_13", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. curtain\nB. window frame\nC. notebook\nD. bedside table", "choices": ["A. curtain", "B. window frame", "C. notebook", "D. bedside table"], "answer": "B. window frame", "target": "phone charger", "split": "train"}
{"id": "4c141d5b1b_14", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the window frame, about 15 cm away. It is also to the side of and below the window, about 21 cm away. It is further to the side of and below the curtain, about 48 cm away.", "target": "phone charger", "split": "train"}
{"id": "4c141d5b1b_15", "scene_id": "4c141d5b1b", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"window frame\", \"relation\": \"side_below\", \"distance_cm\": 15.2}, {\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 21.2}]}", "target": "phone charger", "split": "train"}
{"id": "4c5c60fa76_0", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "4c5c60fa76_1", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. storage cabinet\nB. tripod\nC. board\nD. machine", "choices": ["A. storage cabinet", "B. tripod", "C. board", "D. machine"], "answer": "D. machine", "target": "backpack", "split": "train"}
{"id": "4c5c60fa76_2", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the cabinet. It is also to the side of and below the machine, about 8 cm away. It is further to the side of and above the board, about 16 cm away.", "target": "backpack", "split": "train"}
{"id": "4c5c60fa76_3", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"machine\", \"relation\": \"side_below\", \"distance_cm\": 8.4}, {\"object\": \"board\", \"relation\": \"side_above\", \"distance_cm\": 15.7}]}", "target": "backpack", "split": "train"}
{"id": "4c5c60fa76_4", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "Yes", "target": "calculator", "split": "train"}
{"id": "4c5c60fa76_5", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the calculator in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The calculator is on the table. It is also next to the keyboard, about 3 cm away. It is further to the side of and above the office chair, about 3 cm away.", "target": "calculator", "split": "train"}
{"id": "4c5c60fa76_6", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the calculator in structured JSON format.", "choices": null, "answer": "{\"target\": \"calculator\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 3.5}]}", "target": "calculator", "split": "train"}
{"id": "4c5c60fa76_7", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "4c5c60fa76_8", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is next to the mouse, about 7 cm away. It is also to the side of and above the cabinet, about 7 cm away. It is further to the side of and above the office chair, about 9 cm away.", "target": "scissors", "split": "train"}
{"id": "4c5c60fa76_9", "scene_id": "4c5c60fa76", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 7.4}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 7.4}]}", "target": "scissors", "split": "train"}
{"id": "4d451d9c36_0", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "4d451d9c36_1", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the box, about 10 cm away. It is further to the side of and below the air purifier, about 10 cm away.", "target": "bottle", "split": "train"}
{"id": "4d451d9c36_2", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 9.8}, {\"object\": \"air purifier\", \"relation\": \"side_below\", \"distance_cm\": 10.3}]}", "target": "bottle", "split": "train"}
{"id": "4d451d9c36_3", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "4d451d9c36_4", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse, about 8 cm away. It is also to the side of and below the monitor, about 30 cm away.", "target": "keyboard", "split": "train"}
{"id": "4d451d9c36_5", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"electrical equipment\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 8.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 29.5}]}", "target": "keyboard", "split": "train"}
{"id": "4d451d9c36_6", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "4d451d9c36_7", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. air purifier\nC. keyboard\nD. electrical equipment", "choices": ["A. monitor", "B. air purifier", "C. keyboard", "D. electrical equipment"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "4d451d9c36_8", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 8 cm away. It is also to the side of and below the monitor, about 40 cm away.", "target": "mouse", "split": "train"}
{"id": "4d451d9c36_9", "scene_id": "4d451d9c36", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"electrical equipment\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 39.8}]}", "target": "mouse", "split": "train"}
{"id": "4e0b8cbd33_0", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "4e0b8cbd33_1", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. water bottle\nB. office chair\nC. cabinet\nD. beverage can", "choices": ["A. water bottle", "B. office chair", "C. cabinet", "D. beverage can"], "answer": "C. cabinet", "target": "backpack", "split": "train"}
{"id": "4e0b8cbd33_2", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the cabinet, about 11 cm away. It is also to the side of and below the office chair, about 39 cm away. It is further to the side of and below the water bottle, about 77 cm away.", "target": "backpack", "split": "train"}
{"id": "4e0b8cbd33_3", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 11.1}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 38.6}]}", "target": "backpack", "split": "train"}
{"id": "4e0b8cbd33_4", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "4e0b8cbd33_5", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. monitor\nB. note\nC. box\nD. table", "choices": ["A. monitor", "B. note", "C. box", "D. table"], "answer": "D. table", "target": "phone charger", "split": "train"}
{"id": "4e0b8cbd33_6", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and above the table, about 3 cm away. It is also next to the note, about 14 cm away. It is further to the side of and below the monitor, about 14 cm away.", "target": "phone charger", "split": "train"}
{"id": "4e0b8cbd33_7", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"note\", \"relation\": \"next_to\", \"distance_cm\": 13.7}]}", "target": "phone charger", "split": "train"}
{"id": "4e0b8cbd33_8", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "4e0b8cbd33_9", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. pc\nB. table\nC. mouse\nD. cabinet", "choices": ["A. pc", "B. table", "C. mouse", "D. cabinet"], "answer": "A. pc", "target": "power strip", "split": "train"}
{"id": "4e0b8cbd33_10", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also to the side of and below the pc, about 6 cm away. It is further near the mouse, about 25 cm away.", "target": "power strip", "split": "train"}
{"id": "4e0b8cbd33_11", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pc\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 24.6}]}", "target": "power strip", "split": "train"}
{"id": "4e0b8cbd33_12", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "Yes", "target": "scissors", "split": "train"}
{"id": "4e0b8cbd33_13", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scissors in this scene?\nA. cardboard box\nB. papers\nC. speakers\nD. keyboard", "choices": ["A. cardboard box", "B. papers", "C. speakers", "D. keyboard"], "answer": "A. cardboard box", "target": "scissors", "split": "train"}
{"id": "4e0b8cbd33_14", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is to the side of and below the cardboard box, about 11 cm away. It is also to the side of and below the papers, about 13 cm away. It is further near the keyboard, about 24 cm away.", "target": "scissors", "split": "train"}
{"id": "4e0b8cbd33_15", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cardboard box\", \"relation\": \"side_below\", \"distance_cm\": 11.2}, {\"object\": \"papers\", \"relation\": \"side_below\", \"distance_cm\": 13.3}]}", "target": "scissors", "split": "train"}
{"id": "4e0b8cbd33_16", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "4e0b8cbd33_17", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. pc\nB. reusable water bottle\nC. mug\nD. mouse", "choices": ["A. pc", "B. reusable water bottle", "C. mug", "D. mouse"], "answer": "C. mug", "target": "spray bottle", "split": "train"}
{"id": "4e0b8cbd33_18", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the table. It is also to the side of and above the mug, about 11 cm away. It is further near the reusable water bottle, about 20 cm away.", "target": "spray bottle", "split": "train"}
{"id": "4e0b8cbd33_19", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mug\", \"relation\": \"side_above\", \"distance_cm\": 11.3}, {\"object\": \"reusable water bottle\", \"relation\": \"near\", \"distance_cm\": 20.0}]}", "target": "spray bottle", "split": "train"}
{"id": "4e0b8cbd33_20", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "Yes", "target": "stapler", "split": "train"}
{"id": "4e0b8cbd33_21", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. printer\nB. umbrella\nC. jacket\nD. table", "choices": ["A. printer", "B. umbrella", "C. jacket", "D. table"], "answer": "D. table", "target": "stapler", "split": "train"}
{"id": "4e0b8cbd33_22", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is to the side of and above the table, about 5 cm away. It is also to the side of and below the printer, about 9 cm away. It is further to the side of and below the jacket, about 23 cm away.", "target": "stapler", "split": "train"}
{"id": "4e0b8cbd33_23", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.8}, {\"object\": \"printer\", \"relation\": \"side_below\", \"distance_cm\": 9.2}]}", "target": "stapler", "split": "train"}
{"id": "4e0b8cbd33_24", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "4e0b8cbd33_25", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. computer tower\nB. plant pot\nC. plant\nD. beverage can", "choices": ["A. computer tower", "B. plant pot", "C. plant", "D. beverage can"], "answer": "B. plant pot", "target": "water bottle", "split": "train"}
{"id": "4e0b8cbd33_26", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also next to the plant pot, about 10 cm away. It is further to the side of and above the computer tower, about 17 cm away.", "target": "water bottle", "split": "train"}
{"id": "4e0b8cbd33_27", "scene_id": "4e0b8cbd33", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"plant pot\", \"relation\": \"next_to\", \"distance_cm\": 10.2}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 16.9}]}", "target": "water bottle", "split": "train"}
{"id": "4e9ab3ec88_0", "scene_id": "4e9ab3ec88", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "4e9ab3ec88_1", "scene_id": "4e9ab3ec88", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is above the storage rack. It is also next to the shopping bag, about 6 cm away. It is further to the side of and below the board, about 19 cm away.", "target": "glasses case", "split": "train"}
{"id": "4e9ab3ec88_2", "scene_id": "4e9ab3ec88", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"storage rack\", \"relation\": \"above\", \"distance_cm\": 2.3}, {\"object\": \"shopping bag\", \"relation\": \"next_to\", \"distance_cm\": 6.2}]}", "target": "glasses case", "split": "train"}
{"id": "4e9ab3ec88_3", "scene_id": "4e9ab3ec88", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "4e9ab3ec88_4", "scene_id": "4e9ab3ec88", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. table\nB. table lamp\nC. laptop\nD. chair", "choices": ["A. table", "B. table lamp", "C. laptop", "D. chair"], "answer": "A. table", "target": "headphones", "split": "train"}
{"id": "4e9ab3ec88_5", "scene_id": "4e9ab3ec88", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the table, about 4 cm away. It is also to the side of and above the chair, about 11 cm away. It is further to the side of and below the table lamp, about 25 cm away.", "target": "headphones", "split": "train"}
{"id": "4e9ab3ec88_6", "scene_id": "4e9ab3ec88", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 11.1}]}", "target": "headphones", "split": "train"}
{"id": "4f144c7489_0", "scene_id": "4f144c7489", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "4f144c7489_1", "scene_id": "4f144c7489", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power extension in this scene?\nA. air conditioner\nB. bakcpack\nC. table\nD. heater", "choices": ["A. air conditioner", "B. bakcpack", "C. table", "D. heater"], "answer": "C. table", "target": "power extension", "split": "train"}
{"id": "4f144c7489_2", "scene_id": "4f144c7489", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power extension in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power extension is to the side of and below the table, about 13 cm away. It is also to the side of and below the air conditioner, about 16 cm away. It is further to the side of and below the heater, about 40 cm away.", "target": "power extension", "split": "train"}
{"id": "4f144c7489_3", "scene_id": "4f144c7489", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power extension in structured JSON format.", "choices": null, "answer": "{\"target\": \"power extension\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 12.9}, {\"object\": \"air conditioner\", \"relation\": \"side_below\", \"distance_cm\": 16.1}]}", "target": "power extension", "split": "train"}
{"id": "504cf57907_0", "scene_id": "504cf57907", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "504cf57907_1", "scene_id": "504cf57907", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. headboard\nB. pillow\nC. radiator\nD. backpack", "choices": ["A. headboard", "B. pillow", "C. radiator", "D. backpack"], "answer": "A. headboard", "target": "phone charger", "split": "train"}
{"id": "504cf57907_2", "scene_id": "504cf57907", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the headboard, about 7 cm away. It is also to the side of and below the radiator, about 27 cm away. It is further to the side of and below the pillow, about 34 cm away.", "target": "phone charger", "split": "train"}
{"id": "504cf57907_3", "scene_id": "504cf57907", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"wall lamp\", \"references\": [{\"object\": \"headboard\", \"relation\": \"side_below\", \"distance_cm\": 6.6}, {\"object\": \"radiator\", \"relation\": \"side_below\", \"distance_cm\": 27.4}]}", "target": "phone charger", "split": "train"}
{"id": "504cf57907_4", "scene_id": "504cf57907", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "504cf57907_5", "scene_id": "504cf57907", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is on the table. It is also next to the glove, about 5 cm away. It is further next to the laptop, about 5 cm away.", "target": "power adapter", "split": "train"}
{"id": "504cf57907_6", "scene_id": "504cf57907", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"glove\", \"relation\": \"next_to\", \"distance_cm\": 4.8}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 5.0}]}", "target": "power adapter", "split": "train"}
{"id": "50809ea0d8_0", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "50809ea0d8_1", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bottle\nB. cabinet\nC. jacket\nD. poster", "choices": ["A. bottle", "B. cabinet", "C. jacket", "D. poster"], "answer": "D. poster", "target": "backpack", "split": "train"}
{"id": "50809ea0d8_2", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the poster, about 33 cm away.", "target": "backpack", "split": "train"}
{"id": "50809ea0d8_3", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"poster\", \"references\": [{\"object\": \"poster\", \"relation\": \"side_below\", \"distance_cm\": 33.2}]}", "target": "backpack", "split": "train"}
{"id": "50809ea0d8_4", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "50809ea0d8_5", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mouse\nB. keyboard\nC. monitor\nD. cabinet", "choices": ["A. mouse", "B. keyboard", "C. monitor", "D. cabinet"], "answer": "A. mouse", "target": "bottle", "split": "train"}
{"id": "50809ea0d8_6", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also near the mouse, about 16 cm away. It is further near the keyboard, about 23 cm away.", "target": "bottle", "split": "train"}
{"id": "50809ea0d8_7", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 16.4}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 22.6}]}", "target": "bottle", "split": "train"}
{"id": "50809ea0d8_8", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "50809ea0d8_9", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the lanyard in this scene?\nA. cabinet\nB. backpack\nC. bottle\nD. split", "choices": ["A. cabinet", "B. backpack", "C. bottle", "D. split"], "answer": "D. split", "target": "lanyard", "split": "train"}
{"id": "50809ea0d8_10", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the lanyard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The lanyard is to the side of and above the split, about 33 cm away.", "target": "lanyard", "split": "train"}
{"id": "50809ea0d8_11", "scene_id": "50809ea0d8", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the lanyard in structured JSON format.", "choices": null, "answer": "{\"target\": \"lanyard\", \"support_surface\": \"paper\", \"references\": [{\"object\": \"split\", \"relation\": \"side_above\", \"distance_cm\": 32.7}]}", "target": "lanyard", "split": "train"}
{"id": "5166da6131_0", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "5166da6131_1", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the adapter in this scene?\nA. cable\nB. file cabinet\nC. heater\nD. power socket", "choices": ["A. cable", "B. file cabinet", "C. heater", "D. power socket"], "answer": "A. cable", "target": "adapter", "split": "train"}
{"id": "5166da6131_2", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The adapter is to the side of and above the cable, about 3 cm away. It is also to the side of and above the file cabinet, about 14 cm away. It is further to the side of and above the heater, about 19 cm away.", "target": "adapter", "split": "train"}
{"id": "5166da6131_3", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"adapter\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"cable\", \"relation\": \"side_above\", \"distance_cm\": 2.9}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 13.7}]}", "target": "adapter", "split": "train"}
{"id": "5166da6131_4", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "5166da6131_5", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothes\nB. table\nC. binder\nD. book", "choices": ["A. clothes", "B. table", "C. binder", "D. book"], "answer": "A. clothes", "target": "backpack", "split": "train"}
{"id": "5166da6131_6", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the clothes, about 7 cm away. It is also to the side of and above the table, about 9 cm away. It is further near the binder, about 18 cm away.", "target": "backpack", "split": "train"}
{"id": "5166da6131_7", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_above\", \"distance_cm\": 6.9}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 8.7}]}", "target": "backpack", "split": "train"}
{"id": "5166da6131_8", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "5166da6131_9", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. office chair\nC. cardboard box\nD. monitor", "choices": ["A. mouse", "B. office chair", "C. cardboard box", "D. monitor"], "answer": "A. mouse", "target": "keyboard", "split": "train"}
{"id": "5166da6131_10", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 13 cm away. It is further to the side of and below the cardboard box, about 17 cm away.", "target": "keyboard", "split": "train"}
{"id": "5166da6131_11", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.7}, {\"object\": \"cardboard box\", \"relation\": \"side_below\", \"distance_cm\": 16.9}]}", "target": "keyboard", "split": "train"}
{"id": "5166da6131_12", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "5166da6131_13", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. sofa\nB. chair\nC. papers\nD. box", "choices": ["A. sofa", "B. chair", "C. papers", "D. box"], "answer": "C. papers", "target": "notebook", "split": "train"}
{"id": "5166da6131_14", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also to the side of and above the chair, about 26 cm away. It is further to the side of and above the sofa, about 33 cm away.", "target": "notebook", "split": "train"}
{"id": "5166da6131_15", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 26.3}, {\"object\": \"sofa\", \"relation\": \"side_above\", \"distance_cm\": 32.6}]}", "target": "notebook", "split": "train"}
{"id": "5166da6131_16", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "train"}
{"id": "5166da6131_17", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. book\nB. printer\nC. napkin\nD. bowl", "choices": ["A. book", "B. printer", "C. napkin", "D. bowl"], "answer": "D. bowl", "target": "water bottle", "split": "train"}
{"id": "5166da6131_18", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also to the side of and above the bowl, about 3 cm away. It is further to the side of and above the book, about 11 cm away.", "target": "water bottle", "split": "train"}
{"id": "5166da6131_19", "scene_id": "5166da6131", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bowl\", \"relation\": \"side_above\", \"distance_cm\": 2.6}, {\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 10.7}]}", "target": "water bottle", "split": "train"}
{"id": "51bdbf173f_0", "scene_id": "51bdbf173f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "51bdbf173f_1", "scene_id": "51bdbf173f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the basket. It is also under the storage rack, about 3 cm away. It is further next to the coffee maker, about 6 cm away.", "target": "bottle", "split": "train"}
{"id": "51bdbf173f_2", "scene_id": "51bdbf173f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"side table\", \"references\": [{\"object\": \"basket\", \"relation\": \"side_above\", \"distance_cm\": 2.4}, {\"object\": \"storage rack\", \"relation\": \"under\", \"distance_cm\": 3.0}]}", "target": "bottle", "split": "train"}
{"id": "51dc4178bd_0", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "51dc4178bd_1", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. mini fridge\nB. hair comb\nC. bread\nD. magazines", "choices": ["A. mini fridge", "B. hair comb", "C. bread", "D. magazines"], "answer": "A. mini fridge", "target": "backpack", "split": "train"}
{"id": "51dc4178bd_2", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the mini fridge, about 13 cm away. It is also to the side of and below the bread, about 41 cm away. It is further to the side of and below the magazines, about 42 cm away.", "target": "backpack", "split": "train"}
{"id": "51dc4178bd_3", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"mini fridge\", \"relation\": \"next_to\", \"distance_cm\": 13.5}, {\"object\": \"bread\", \"relation\": \"side_below\", \"distance_cm\": 40.9}]}", "target": "backpack", "split": "train"}
{"id": "51dc4178bd_4", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "51dc4178bd_5", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. power sockets\nB. chair\nC. book\nD. pedestal fan", "choices": ["A. power sockets", "B. chair", "C. book", "D. pedestal fan"], "answer": "C. book", "target": "extension cord", "split": "train"}
{"id": "51dc4178bd_6", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is near the book, about 30 cm away. It is also to the side of and above the chair, about 37 cm away. It is further near the pedestal fan, about 40 cm away.", "target": "extension cord", "split": "train"}
{"id": "51dc4178bd_7", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 30.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 36.9}]}", "target": "extension cord", "split": "train"}
{"id": "51dc4178bd_8", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "51dc4178bd_9", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. toilet paper roll\nB. ketchup bottle\nC. book\nD. chair", "choices": ["A. toilet paper roll", "B. ketchup bottle", "C. book", "D. chair"], "answer": "D. chair", "target": "notebook", "split": "train"}
{"id": "51dc4178bd_10", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and above the chair, about 9 cm away. It is also to the side of and below the ketchup bottle, about 16 cm away. It is further near the toilet paper roll, about 18 cm away.", "target": "notebook", "split": "train"}
{"id": "51dc4178bd_11", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 9.4}, {\"object\": \"ketchup bottle\", \"relation\": \"side_below\", \"distance_cm\": 16.1}]}", "target": "notebook", "split": "train"}
{"id": "51dc4178bd_12", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "51dc4178bd_13", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. bedside table\nB. light switch\nC. backpack\nD. pillow", "choices": ["A. bedside table", "B. light switch", "C. backpack", "D. pillow"], "answer": "B. light switch", "target": "phone charger", "split": "train"}
{"id": "51dc4178bd_14", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is under the light switch, about 3 cm away. It is also near the pillow, about 30 cm away.", "target": "phone charger", "split": "train"}
{"id": "51dc4178bd_15", "scene_id": "51dc4178bd", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"light switch\", \"relation\": \"under\", \"distance_cm\": 3.1}, {\"object\": \"pillow\", \"relation\": \"near\", \"distance_cm\": 29.8}]}", "target": "phone charger", "split": "train"}
{"id": "523657b4d0_0", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "523657b4d0_1", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. table\nB. whiteboard eraser\nC. power sockets\nD. light switches", "choices": ["A. table", "B. whiteboard eraser", "C. power sockets", "D. light switches"], "answer": "A. table", "target": "power adapter", "split": "train"}
{"id": "523657b4d0_2", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and above the table, about 9 cm away. It is also near the light switches, about 22 cm away. It is further to the side of and below the whiteboard eraser, about 32 cm away.", "target": "power adapter", "split": "train"}
{"id": "523657b4d0_3", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 8.9}, {\"object\": \"light switches\", \"relation\": \"near\", \"distance_cm\": 21.9}]}", "target": "power adapter", "split": "train"}
{"id": "523657b4d0_4", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "523657b4d0_5", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote control in this scene?\nA. power adapter\nB. light switches\nC. chair\nD. power sockets", "choices": ["A. power adapter", "B. light switches", "C. chair", "D. power sockets"], "answer": "C. chair", "target": "remote control", "split": "train"}
{"id": "523657b4d0_6", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote control in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote control is on the table. It is also to the side of and above the chair, about 14 cm away. It is further to the side of and below the light switches, about 39 cm away.", "target": "remote control", "split": "train"}
{"id": "523657b4d0_7", "scene_id": "523657b4d0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote control in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote control\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 13.8}, {\"object\": \"light switches\", \"relation\": \"side_below\", \"distance_cm\": 38.7}]}", "target": "remote control", "split": "train"}
{"id": "52599ae063_0", "scene_id": "52599ae063", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "52599ae063_1", "scene_id": "52599ae063", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the trash can, about 78 cm away. It is also to the side of and below the light switch, about 79 cm away.", "target": "bottle", "split": "train"}
{"id": "52599ae063_2", "scene_id": "52599ae063", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 77.8}, {\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 78.6}]}", "target": "bottle", "split": "train"}
{"id": "52599ae063_3", "scene_id": "52599ae063", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "Yes", "target": "post it note", "split": "train"}
{"id": "52599ae063_4", "scene_id": "52599ae063", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the post it note in this scene?\nA. air purifier\nB. bottle\nC. cabinet\nD. bookshelf", "choices": ["A. air purifier", "B. bottle", "C. cabinet", "D. bookshelf"], "answer": "C. cabinet", "target": "post it note", "split": "train"}
{"id": "52599ae063_5", "scene_id": "52599ae063", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the post it note in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The post it note is to the side of and above the cabinet, about 60 cm away. It is also to the side of and above the air purifier, about 74 cm away.", "target": "post it note", "split": "train"}
{"id": "52599ae063_6", "scene_id": "52599ae063", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the post it note in structured JSON format.", "choices": null, "answer": "{\"target\": \"post it note\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 59.8}, {\"object\": \"air purifier\", \"relation\": \"side_above\", \"distance_cm\": 73.8}]}", "target": "post it note", "split": "train"}
{"id": "5334a4164a_0", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "5334a4164a_1", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. baseball cap\nB. bench\nC. blanket\nD. basket", "choices": ["A. baseball cap", "B. bench", "C. blanket", "D. basket"], "answer": "C. blanket", "target": "backpack", "split": "train"}
{"id": "5334a4164a_2", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the carpet. It is also to the side of and below the blanket, about 21 cm away.", "target": "backpack", "split": "train"}
{"id": "5334a4164a_3", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"carpet\", \"references\": [{\"object\": \"blanket\", \"relation\": \"side_below\", \"distance_cm\": 20.6}]}", "target": "backpack", "split": "train"}
{"id": "5334a4164a_4", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "5334a4164a_5", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. power sockets\nB. jacket\nC. paper\nD. notebook", "choices": ["A. power sockets", "B. jacket", "C. paper", "D. notebook"], "answer": "A. power sockets", "target": "extension cord", "split": "train"}
{"id": "5334a4164a_6", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and above the power sockets, about 6 cm away. It is also to the side of and below the paper, about 7 cm away. It is further to the side of and below the notebook, about 9 cm away.", "target": "extension cord", "split": "train"}
{"id": "5334a4164a_7", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"power sockets\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"paper\", \"relation\": \"side_below\", \"distance_cm\": 6.8}]}", "target": "extension cord", "split": "train"}
{"id": "5334a4164a_8", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "5334a4164a_9", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the handbag in this scene?\nA. baseball cap\nB. backpack\nC. jackets\nD. trolley", "choices": ["A. baseball cap", "B. backpack", "C. jackets", "D. trolley"], "answer": "A. baseball cap", "target": "handbag", "split": "train"}
{"id": "5334a4164a_10", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the handbag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The handbag is on the cabinet. It is also to the side of and below the baseball cap, about 13 cm away. It is further near the jackets, about 21 cm away.", "target": "handbag", "split": "train"}
{"id": "5334a4164a_11", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the handbag in structured JSON format.", "choices": null, "answer": "{\"target\": \"handbag\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"baseball cap\", \"relation\": \"side_below\", \"distance_cm\": 13.1}, {\"object\": \"jackets\", \"relation\": \"near\", \"distance_cm\": 21.4}]}", "target": "handbag", "split": "train"}
{"id": "5334a4164a_12", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "5334a4164a_13", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. notebook\nB. paper\nC. desk\nD. jacket", "choices": ["A. notebook", "B. paper", "C. desk", "D. jacket"], "answer": "C. desk", "target": "phone charger", "split": "train"}
{"id": "5334a4164a_14", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and above the desk, about 6 cm away. It is also below the paper, about 10 cm away. It is further to the side of and above the notebook, about 16 cm away.", "target": "phone charger", "split": "train"}
{"id": "5334a4164a_15", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 6.2}, {\"object\": \"paper\", \"relation\": \"below\", \"distance_cm\": 9.9}]}", "target": "phone charger", "split": "train"}
{"id": "5334a4164a_16", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "5334a4164a_17", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. blanket\nB. books\nC. power sockets\nD. basket", "choices": ["A. blanket", "B. books", "C. power sockets", "D. basket"], "answer": "C. power sockets", "target": "remote", "split": "train"}
{"id": "5334a4164a_18", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is to the side of and below the power sockets, about 11 cm away. It is also near the blanket, about 23 cm away. It is further below the basket, about 29 cm away.", "target": "remote", "split": "train"}
{"id": "5334a4164a_19", "scene_id": "5334a4164a", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 10.8}, {\"object\": \"blanket\", \"relation\": \"near\", \"distance_cm\": 22.8}]}", "target": "remote", "split": "train"}
{"id": "53bd1b3b05_0", "scene_id": "53bd1b3b05", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "53bd1b3b05_1", "scene_id": "53bd1b3b05", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. junction box\nB. doorframe\nC. monitor\nD. door stopper", "choices": ["A. junction box", "B. doorframe", "C. monitor", "D. door stopper"], "answer": "B. doorframe", "target": "cable", "split": "train"}
{"id": "53bd1b3b05_2", "scene_id": "53bd1b3b05", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and above the doorframe, about 15 cm away. It is also to the side of and below the door stopper, about 52 cm away.", "target": "cable", "split": "train"}
{"id": "53bd1b3b05_3", "scene_id": "53bd1b3b05", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"door stopper\", \"references\": [{\"object\": \"doorframe\", \"relation\": \"side_above\", \"distance_cm\": 14.8}, {\"object\": \"door stopper\", \"relation\": \"side_below\", \"distance_cm\": 51.9}]}", "target": "cable", "split": "train"}
{"id": "546292a9db_0", "scene_id": "546292a9db", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "546292a9db_1", "scene_id": "546292a9db", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. tool\nB. cabinet\nC. cloth\nD. lamp", "choices": ["A. tool", "B. cabinet", "C. cloth", "D. lamp"], "answer": "A. tool", "target": "notebook", "split": "train"}
{"id": "546292a9db_2", "scene_id": "546292a9db", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the desk. It is also next to the tool, about 5 cm away. It is further to the side of and above the cloth, about 13 cm away.", "target": "notebook", "split": "train"}
{"id": "546292a9db_3", "scene_id": "546292a9db", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"tool\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"cloth\", \"relation\": \"side_above\", \"distance_cm\": 12.8}]}", "target": "notebook", "split": "train"}
{"id": "55b2bf8036_0", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "55b2bf8036_1", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. keyboard\nB. monitor\nC. cabinet\nD. office chair", "choices": ["A. keyboard", "B. monitor", "C. cabinet", "D. office chair"], "answer": "A. keyboard", "target": "bottle", "split": "train"}
{"id": "55b2bf8036_2", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the keyboard, about 3 cm away. It is also to the side of and below the monitor, about 8 cm away. It is further to the side of and above the cabinet, about 24 cm away.", "target": "bottle", "split": "train"}
{"id": "55b2bf8036_3", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.9}]}", "target": "bottle", "split": "train"}
{"id": "55b2bf8036_4", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "55b2bf8036_5", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. office table\nB. monitor\nC. pencil holder\nD. storage box", "choices": ["A. office table", "B. monitor", "C. pencil holder", "D. storage box"], "answer": "B. monitor", "target": "cup", "split": "train"}
{"id": "55b2bf8036_6", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is under the monitor, about 4 cm away. It is also to the side of and above the office table, about 12 cm away. It is further to the side of and above the storage box, about 19 cm away.", "target": "cup", "split": "train"}
{"id": "55b2bf8036_7", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"monitor stand\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.9}, {\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 11.9}]}", "target": "cup", "split": "train"}
{"id": "55b2bf8036_8", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "No", "target": "highlighter", "split": "train"}
{"id": "55b2bf8036_9", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. office chair\nB. laptop stand\nC. keyboard\nD. monitor", "choices": ["A. office chair", "B. laptop stand", "C. keyboard", "D. monitor"], "answer": "C. keyboard", "target": "headphones", "split": "train"}
{"id": "55b2bf8036_10", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is near the keyboard, about 18 cm away. It is also near the laptop stand, about 22 cm away. It is further to the side of and above the office chair, about 26 cm away.", "target": "headphones", "split": "train"}
{"id": "55b2bf8036_11", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 18.1}, {\"object\": \"laptop stand\", \"relation\": \"near\", \"distance_cm\": 21.8}]}", "target": "headphones", "split": "train"}
{"id": "55b2bf8036_12", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "Yes", "target": "pencil holder", "split": "train"}
{"id": "55b2bf8036_13", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil holder in this scene?\nA. monitor stand\nB. monitor\nC. office table\nD. cup", "choices": ["A. monitor stand", "B. monitor", "C. office table", "D. cup"], "answer": "B. monitor", "target": "pencil holder", "split": "train"}
{"id": "55b2bf8036_14", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil holder is to the side of and below the monitor, about 16 cm away. It is also to the side of and above the office table, about 19 cm away. It is further to the side of and above the monitor stand, about 29 cm away.", "target": "pencil holder", "split": "train"}
{"id": "55b2bf8036_15", "scene_id": "55b2bf8036", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil holder\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 15.9}, {\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 18.5}]}", "target": "pencil holder", "split": "train"}
{"id": "56a0ec536c_0", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "56a0ec536c_1", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. box\nB. sticky notes\nC. monitor\nD. keyboard", "choices": ["A. box", "B. sticky notes", "C. monitor", "D. keyboard"], "answer": "B. sticky notes", "target": "bottle", "split": "train"}
{"id": "56a0ec536c_2", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the sticky notes, about 3 cm away. It is also to the side of and below the monitor, about 6 cm away. It is further to the side of and above the keyboard, about 12 cm away.", "target": "bottle", "split": "train"}
{"id": "56a0ec536c_3", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"sticky notes\", \"relation\": \"side_above\", \"distance_cm\": 3.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.5}]}", "target": "bottle", "split": "train"}
{"id": "56a0ec536c_4", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "56a0ec536c_5", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. paper\nB. stack of paper\nC. split\nD. office table", "choices": ["A. paper", "B. stack of paper", "C. split", "D. office table"], "answer": "B. stack of paper", "target": "keyboard", "split": "train"}
{"id": "56a0ec536c_6", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is near the stack of paper, about 18 cm away. It is also near the paper, about 19 cm away. It is further near the split, about 31 cm away.", "target": "keyboard", "split": "train"}
{"id": "56a0ec536c_7", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"stack of paper\", \"relation\": \"near\", \"distance_cm\": 17.5}, {\"object\": \"paper\", \"relation\": \"near\", \"distance_cm\": 19.4}]}", "target": "keyboard", "split": "train"}
{"id": "56a0ec536c_8", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "Yes", "target": "stapler", "split": "train"}
{"id": "56a0ec536c_9", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. office table\nB. cables\nC. box\nD. monitor", "choices": ["A. office table", "B. cables", "C. box", "D. monitor"], "answer": "A. office table", "target": "stapler", "split": "train"}
{"id": "56a0ec536c_10", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is to the side of and above the office table, about 11 cm away. It is also to the side of and below the monitor, about 12 cm away. It is further near the cables, about 21 cm away.", "target": "stapler", "split": "train"}
{"id": "56a0ec536c_11", "scene_id": "56a0ec536c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 11.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.4}]}", "target": "stapler", "split": "train"}
{"id": "578511c8a9_0", "scene_id": "578511c8a9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the deodorant bottle appear?", "choices": null, "answer": "No", "target": "deodorant bottle", "split": "train"}
{"id": "578511c8a9_1", "scene_id": "578511c8a9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. office chair\nB. books\nC. jug\nD. cabinet", "choices": ["A. office chair", "B. books", "C. jug", "D. cabinet"], "answer": "C. jug", "target": "mug", "split": "train"}
{"id": "578511c8a9_2", "scene_id": "578511c8a9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and below the jug, about 8 cm away. It is further to the side of and above the office chair, about 15 cm away.", "target": "mug", "split": "train"}
{"id": "578511c8a9_3", "scene_id": "578511c8a9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"jug\", \"relation\": \"side_below\", \"distance_cm\": 8.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 14.6}]}", "target": "mug", "split": "train"}
{"id": "58960ff105_0", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "58960ff105_1", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is near the electric kettle, about 25 cm away. It is also to the side of and above the pot, about 26 cm away. It is further to the side of and above the mug, about 27 cm away.", "target": "bottle", "split": "train"}
{"id": "58960ff105_2", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"electric kettle\", \"relation\": \"near\", \"distance_cm\": 24.8}, {\"object\": \"pot\", \"relation\": \"side_above\", \"distance_cm\": 25.6}]}", "target": "bottle", "split": "train"}
{"id": "58960ff105_3", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the comb appear?", "choices": null, "answer": "No", "target": "comb", "split": "train"}
{"id": "58960ff105_4", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. kitchen cabinet\nB. serving tray\nC. mug\nD. dishes", "choices": ["A. kitchen cabinet", "B. serving tray", "C. mug", "D. dishes"], "answer": "B. serving tray", "target": "cup", "split": "train"}
{"id": "58960ff105_5", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the serving tray, about 13 cm away. It is also to the side of and above the dishes, about 19 cm away. It is further to the side of and above the kitchen cabinet, about 27 cm away.", "target": "cup", "split": "train"}
{"id": "58960ff105_6", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"serving tray\", \"references\": [{\"object\": \"serving tray\", \"relation\": \"side_below\", \"distance_cm\": 13.4}, {\"object\": \"dishes\", \"relation\": \"side_above\", \"distance_cm\": 18.6}]}", "target": "cup", "split": "train"}
{"id": "58960ff105_7", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "58960ff105_8", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. serving tray\nB. bottle\nC. sink\nD. electric kettle", "choices": ["A. serving tray", "B. bottle", "C. sink", "D. electric kettle"], "answer": "D. electric kettle", "target": "mug", "split": "train"}
{"id": "58960ff105_9", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and below the electric kettle, about 7 cm away. It is also to the side of and above the sink, about 13 cm away. It is further to the side of and below the serving tray, about 24 cm away.", "target": "mug", "split": "train"}
{"id": "58960ff105_10", "scene_id": "58960ff105", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"electric kettle\", \"relation\": \"side_below\", \"distance_cm\": 6.8}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 13.4}]}", "target": "mug", "split": "train"}
{"id": "58f6a5c5ec_0", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "58f6a5c5ec_1", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. laptop charger\nB. laptop\nC. tablet\nD. chair", "choices": ["A. laptop charger", "B. laptop", "C. tablet", "D. chair"], "answer": "C. tablet", "target": "headphones", "split": "train"}
{"id": "58f6a5c5ec_2", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also near the tablet, about 26 cm away. It is further to the side of and above the chair, about 32 cm away.", "target": "headphones", "split": "train"}
{"id": "58f6a5c5ec_3", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"tablet\", \"relation\": \"near\", \"distance_cm\": 26.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 32.4}]}", "target": "headphones", "split": "train"}
{"id": "58f6a5c5ec_4", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "58f6a5c5ec_5", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. headphones\nB. chair\nC. tablet\nD. laptop", "choices": ["A. headphones", "B. chair", "C. tablet", "D. laptop"], "answer": "D. laptop", "target": "laptop charger", "split": "train"}
{"id": "58f6a5c5ec_6", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is near the laptop, about 34 cm away. It is also near the tablet, about 39 cm away. It is further near the headphones, about 43 cm away.", "target": "laptop charger", "split": "train"}
{"id": "58f6a5c5ec_7", "scene_id": "58f6a5c5ec", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 34.0}, {\"object\": \"tablet\", \"relation\": \"near\", \"distance_cm\": 38.6}]}", "target": "laptop charger", "split": "train"}
{"id": "59e3f1ea37_0", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "59e3f1ea37_1", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. box\nB. jar\nC. notepad\nD. notebook", "choices": ["A. box", "B. jar", "C. notepad", "D. notebook"], "answer": "C. notepad", "target": "cup", "split": "train"}
{"id": "59e3f1ea37_2", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the notepad, about 12 cm away. It is also near the jar, about 20 cm away. It is further near the box, about 22 cm away.", "target": "cup", "split": "train"}
{"id": "59e3f1ea37_3", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"office desk\", \"references\": [{\"object\": \"notepad\", \"relation\": \"next_to\", \"distance_cm\": 12.5}, {\"object\": \"jar\", \"relation\": \"near\", \"distance_cm\": 20.0}]}", "target": "cup", "split": "train"}
{"id": "59e3f1ea37_4", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "59e3f1ea37_5", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. cabinet\nB. stapler\nC. notebook\nD. telephone", "choices": ["A. cabinet", "B. stapler", "C. notebook", "D. telephone"], "answer": "C. notebook", "target": "headphones", "split": "train"}
{"id": "59e3f1ea37_6", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the notebook. It is also next to the stapler, about 15 cm away. It is further near the telephone, about 17 cm away.", "target": "headphones", "split": "train"}
{"id": "59e3f1ea37_7", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"stapler\", \"relation\": \"next_to\", \"distance_cm\": 15.0}]}", "target": "headphones", "split": "train"}
{"id": "59e3f1ea37_8", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "59e3f1ea37_9", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. notebook\nB. office chair\nC. monitor\nD. bottle", "choices": ["A. notebook", "B. office chair", "C. monitor", "D. bottle"], "answer": "D. bottle", "target": "keyboard", "split": "train"}
{"id": "59e3f1ea37_10", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the bottle, about 16 cm away. It is also near the notebook, about 19 cm away. It is further to the side of and above the office chair, about 21 cm away.", "target": "keyboard", "split": "train"}
{"id": "59e3f1ea37_11", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office desk\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 15.9}, {\"object\": \"notebook\", \"relation\": \"near\", \"distance_cm\": 19.4}]}", "target": "keyboard", "split": "train"}
{"id": "59e3f1ea37_12", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "59e3f1ea37_13", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. jar\nB. bottle\nC. keyboard\nD. office chair", "choices": ["A. jar", "B. bottle", "C. keyboard", "D. office chair"], "answer": "B. bottle", "target": "notebook", "split": "train"}
{"id": "59e3f1ea37_14", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is near the keyboard, about 19 cm away. It is also to the side of and above the office chair, about 38 cm away. It is further near the jar, about 45 cm away.", "target": "notebook", "split": "train"}
{"id": "59e3f1ea37_15", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"office desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 19.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 37.9}]}", "target": "notebook", "split": "train"}
{"id": "59e3f1ea37_16", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "Yes", "target": "notepad", "split": "train"}
{"id": "59e3f1ea37_17", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notepad in this scene?\nA. cabinet\nB. notebook\nC. stapler\nD. telephone", "choices": ["A. cabinet", "B. notebook", "C. stapler", "D. telephone"], "answer": "B. notebook", "target": "notepad", "split": "train"}
{"id": "59e3f1ea37_18", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notepad is next to the notebook, about 8 cm away. It is also near the telephone, about 15 cm away. It is further near the stapler, about 17 cm away.", "target": "notepad", "split": "train"}
{"id": "59e3f1ea37_19", "scene_id": "59e3f1ea37", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"notepad\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 7.7}, {\"object\": \"telephone\", \"relation\": \"near\", \"distance_cm\": 15.0}]}", "target": "notepad", "split": "train"}
{"id": "5a269ba6fe_0", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "5a269ba6fe_1", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil in this scene?\nA. book\nB. power socket\nC. box\nD. chair", "choices": ["A. book", "B. power socket", "C. box", "D. chair"], "answer": "B. power socket", "target": "pencil", "split": "train"}
{"id": "5a269ba6fe_2", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil is to the side of and above the power socket, about 3 cm away.", "target": "pencil", "split": "train"}
{"id": "5a269ba6fe_3", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil\", \"support_surface\": \"power socket\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 3.4}]}", "target": "pencil", "split": "train"}
{"id": "5a269ba6fe_4", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "5a269ba6fe_5", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. table\nB. chair\nC. power socket\nD. heater", "choices": ["A. table", "B. chair", "C. power socket", "D. heater"], "answer": "A. table", "target": "power strip", "split": "train"}
{"id": "5a269ba6fe_6", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the table, about 9 cm away. It is also to the side of and below the chair, about 26 cm away. It is further to the side of and below the heater, about 30 cm away.", "target": "power strip", "split": "train"}
{"id": "5a269ba6fe_7", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 9.1}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 25.6}]}", "target": "power strip", "split": "train"}
{"id": "5a269ba6fe_8", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "Yes", "target": "remote control", "split": "train"}
{"id": "5a269ba6fe_9", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote control in this scene?\nA. box\nB. power socket\nC. chair\nD. book", "choices": ["A. box", "B. power socket", "C. chair", "D. book"], "answer": "A. box", "target": "remote control", "split": "train"}
{"id": "5a269ba6fe_10", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote control in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote control is to the side of and above the box, about 5 cm away. It is also to the side of and below the power socket, about 65 cm away.", "target": "remote control", "split": "train"}
{"id": "5a269ba6fe_11", "scene_id": "5a269ba6fe", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote control in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote control\", \"support_surface\": \"box\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 5.3}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 65.2}]}", "target": "remote control", "split": "train"}
{"id": "5a9cdde1ba_0", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "5a9cdde1ba_1", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. bottle\nB. power socket\nC. bucket\nD. box", "choices": ["A. bottle", "B. power socket", "C. bucket", "D. box"], "answer": "C. bucket", "target": "mug", "split": "train"}
{"id": "5a9cdde1ba_2", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is under the bucket, about 4 cm away. It is also to the side of and below the bottle, about 7 cm away. It is further near the box, about 19 cm away.", "target": "mug", "split": "train"}
{"id": "5a9cdde1ba_3", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"bucket\", \"relation\": \"under\", \"distance_cm\": 4.1}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 7.2}]}", "target": "mug", "split": "train"}
{"id": "5a9cdde1ba_4", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "No", "target": "handbag", "split": "train"}
{"id": "5a9cdde1ba_5", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. backpack\nB. chair\nC. heater\nD. cloth", "choices": ["A. backpack", "B. chair", "C. heater", "D. cloth"], "answer": "D. cloth", "target": "notebook", "split": "train"}
{"id": "5a9cdde1ba_6", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also under the cloth, about 3 cm away. It is further to the side of and above the chair, about 22 cm away.", "target": "notebook", "split": "train"}
{"id": "5a9cdde1ba_7", "scene_id": "5a9cdde1ba", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cloth\", \"relation\": \"under\", \"distance_cm\": 2.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 21.8}]}", "target": "notebook", "split": "train"}
{"id": "5aeac3800a_0", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "5aeac3800a_1", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. plant\nB. notebooks\nC. table lamp\nD. roller blinds", "choices": ["A. plant", "B. notebooks", "C. table lamp", "D. roller blinds"], "answer": "C. table lamp", "target": "pen holder", "split": "train"}
{"id": "5aeac3800a_2", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the desk. It is also under the table lamp. It is further to the side of and below the roller blinds, about 14 cm away.", "target": "pen holder", "split": "train"}
{"id": "5aeac3800a_3", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"under\", \"distance_cm\": 2.2}, {\"object\": \"roller blinds\", \"relation\": \"side_below\", \"distance_cm\": 14.2}]}", "target": "pen holder", "split": "train"}
{"id": "5aeac3800a_4", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "5aeac3800a_5", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. cabinet\nB. pillow\nC. folded blanket\nD. power strip", "choices": ["A. cabinet", "B. pillow", "C. folded blanket", "D. power strip"], "answer": "A. cabinet", "target": "phone charger", "split": "train"}
{"id": "5aeac3800a_6", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is next to the cabinet, about 3 cm away. It is also to the side of and below the pillow, about 5 cm away. It is further to the side of and above the folded blanket, about 24 cm away.", "target": "phone charger", "split": "train"}
{"id": "5aeac3800a_7", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 5.0}]}", "target": "phone charger", "split": "train"}
{"id": "5aeac3800a_8", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "5aeac3800a_9", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. pen holder\nB. plant\nC. roller blinds\nD. utensils", "choices": ["A. pen holder", "B. plant", "C. roller blinds", "D. utensils"], "answer": "C. roller blinds", "target": "stapler", "split": "train"}
{"id": "5aeac3800a_10", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is on the desk. It is also to the side of and below the roller blinds, about 16 cm away. It is further near the plant, about 22 cm away.", "target": "stapler", "split": "train"}
{"id": "5aeac3800a_11", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"roller blinds\", \"relation\": \"side_below\", \"distance_cm\": 16.1}, {\"object\": \"plant\", \"relation\": \"near\", \"distance_cm\": 21.6}]}", "target": "stapler", "split": "train"}
{"id": "5aeac3800a_12", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "5aeac3800a_13", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the wallet in this scene?\nA. table lamp\nB. stapler\nC. office chair\nD. notebooks", "choices": ["A. table lamp", "B. stapler", "C. office chair", "D. notebooks"], "answer": "D. notebooks", "target": "wallet", "split": "train"}
{"id": "5aeac3800a_14", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the wallet in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The wallet is on the desk. It is also to the side of and below the notebooks, about 10 cm away. It is further to the side of and above the office chair, about 15 cm away.", "target": "wallet", "split": "train"}
{"id": "5aeac3800a_15", "scene_id": "5aeac3800a", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the wallet in structured JSON format.", "choices": null, "answer": "{\"target\": \"wallet\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"notebooks\", \"relation\": \"side_below\", \"distance_cm\": 9.8}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 14.8}]}", "target": "wallet", "split": "train"}
{"id": "5bc6227191_0", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "5bc6227191_1", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. toilet paper\nB. books\nC. monitor\nD. laptop", "choices": ["A. toilet paper", "B. books", "C. monitor", "D. laptop"], "answer": "D. laptop", "target": "mouse", "split": "train"}
{"id": "5bc6227191_2", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the laptop, about 6 cm away. It is also to the side of and below the monitor, about 8 cm away. It is further to the side of and below the toilet paper, about 11 cm away.", "target": "mouse", "split": "train"}
{"id": "5bc6227191_3", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.5}]}", "target": "mouse", "split": "train"}
{"id": "5bc6227191_4", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "train"}
{"id": "5bc6227191_5", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. heater\nB. paper\nC. milk carton\nD. water filter", "choices": ["A. heater", "B. paper", "C. milk carton", "D. water filter"], "answer": "B. paper", "target": "water bottle", "split": "train"}
{"id": "5bc6227191_6", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is next to the paper, about 5 cm away. It is also next to the water filter, about 7 cm away. It is further next to the milk carton, about 11 cm away.", "target": "water bottle", "split": "train"}
{"id": "5bc6227191_7", "scene_id": "5bc6227191", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"window sill\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 4.5}, {\"object\": \"water filter\", \"relation\": \"next_to\", \"distance_cm\": 6.5}]}", "target": "water bottle", "split": "train"}
{"id": "5c215ef3b0_0", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the comb appear?", "choices": null, "answer": "No", "target": "comb", "split": "train"}
{"id": "5c215ef3b0_1", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the sponge. It is also to the side of and above the sink. It is further next to the faucet, about 5 cm away.", "target": "cup", "split": "train"}
{"id": "5c215ef3b0_2", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"sponge\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 2.5}]}", "target": "cup", "split": "train"}
{"id": "5c215ef3b0_3", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "5c215ef3b0_4", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. decoration\nB. mug\nC. mouse\nD. folder", "choices": ["A. decoration", "B. mug", "C. mouse", "D. folder"], "answer": "C. mouse", "target": "keyboard", "split": "train"}
{"id": "5c215ef3b0_5", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 6 cm away. It is further next to the mug, about 11 cm away.", "target": "keyboard", "split": "train"}
{"id": "5c215ef3b0_6", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.9}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 11.2}]}", "target": "keyboard", "split": "train"}
{"id": "5c215ef3b0_7", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "5c215ef3b0_8", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. power socket\nB. socks\nC. cardboard box\nD. portable vacuum cleaner", "choices": ["A. power socket", "B. socks", "C. cardboard box", "D. portable vacuum cleaner"], "answer": "D. portable vacuum cleaner", "target": "laptop charger", "split": "train"}
{"id": "5c215ef3b0_9", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the portable vacuum cleaner, about 17 cm away. It is also near the carpet, about 26 cm away. It is further to the side of and below the cardboard box, about 41 cm away.", "target": "laptop charger", "split": "train"}
{"id": "5c215ef3b0_10", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"carpet\", \"references\": [{\"object\": \"portable vacuum cleaner\", \"relation\": \"side_below\", \"distance_cm\": 17.0}, {\"object\": \"carpet\", \"relation\": \"near\", \"distance_cm\": 26.1}]}", "target": "laptop charger", "split": "train"}
{"id": "5c215ef3b0_11", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "5c215ef3b0_12", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. laptop\nB. keyboard\nC. office chair\nD. monitor", "choices": ["A. laptop", "B. keyboard", "C. office chair", "D. monitor"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "5c215ef3b0_13", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 6 cm away. It is further to the side of and below the monitor, about 11 cm away.", "target": "mouse", "split": "train"}
{"id": "5c215ef3b0_14", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.7}]}", "target": "mouse", "split": "train"}
{"id": "5c215ef3b0_15", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "No", "target": "handbag", "split": "train"}
{"id": "5c215ef3b0_16", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. cushion\nB. sofa\nC. extension cord\nD. router", "choices": ["A. cushion", "B. sofa", "C. extension cord", "D. router"], "answer": "B. sofa", "target": "phone charger", "split": "train"}
{"id": "5c215ef3b0_17", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the sofa, about 15 cm away. It is also to the side of and above the router, about 18 cm away. It is further to the side of and above the extension cord, about 39 cm away.", "target": "phone charger", "split": "train"}
{"id": "5c215ef3b0_18", "scene_id": "5c215ef3b0", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"sofa\", \"relation\": \"side_below\", \"distance_cm\": 14.6}, {\"object\": \"router\", \"relation\": \"side_above\", \"distance_cm\": 18.3}]}", "target": "phone charger", "split": "train"}
{"id": "5d591ff74d_0", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "train"}
{"id": "5d591ff74d_1", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cabinet\nB. outlet\nC. split\nD. mouse", "choices": ["A. cabinet", "B. outlet", "C. split", "D. mouse"], "answer": "C. split", "target": "backpack", "split": "train"}
{"id": "5d591ff74d_2", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is near the split, about 16 cm away. It is also to the side of and above the cabinet, about 45 cm away. It is further to the side of and below the mouse, about 65 cm away.", "target": "backpack", "split": "train"}
{"id": "5d591ff74d_3", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"split\", \"relation\": \"near\", \"distance_cm\": 15.7}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 44.7}]}", "target": "backpack", "split": "train"}
{"id": "5d591ff74d_4", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "5d591ff74d_5", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the table, about 10 cm away. It is also to the side of and above the water filter, about 11 cm away. It is further to the side of and above the mouse, about 27 cm away.", "target": "bottle", "split": "train"}
{"id": "5d591ff74d_6", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 10.5}, {\"object\": \"water filter\", \"relation\": \"side_above\", \"distance_cm\": 11.1}]}", "target": "bottle", "split": "train"}
{"id": "5d591ff74d_7", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "5d591ff74d_8", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. paper\nB. office chair\nC. monitor\nD. laptop case", "choices": ["A. paper", "B. office chair", "C. monitor", "D. laptop case"], "answer": "B. office chair", "target": "keyboard", "split": "train"}
{"id": "5d591ff74d_9", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the office chair, about 10 cm away. It is further to the side of and below the monitor, about 35 cm away.", "target": "keyboard", "split": "train"}
{"id": "5d591ff74d_10", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 10.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 35.2}]}", "target": "keyboard", "split": "train"}
{"id": "5d591ff74d_11", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "5d591ff74d_12", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. water filter\nB. office chair\nC. cabinet\nD. bottle", "choices": ["A. water filter", "B. office chair", "C. cabinet", "D. bottle"], "answer": "B. office chair", "target": "mouse", "split": "train"}
{"id": "5d591ff74d_13", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and above the office chair, about 18 cm away. It is further to the side of and above the cabinet, about 26 cm away.", "target": "mouse", "split": "train"}
{"id": "5d591ff74d_14", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 18.0}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 25.9}]}", "target": "mouse", "split": "train"}
{"id": "5d591ff74d_15", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "5d591ff74d_16", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. backpack\nB. box\nC. bookshelf\nD. document holder", "choices": ["A. backpack", "B. box", "C. bookshelf", "D. document holder"], "answer": "B. box", "target": "power strip", "split": "train"}
{"id": "5d591ff74d_17", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and above the box, about 23 cm away. It is also to the side of and below the bookshelf, about 27 cm away. It is further to the side of and below the document holder, about 30 cm away.", "target": "power strip", "split": "train"}
{"id": "5d591ff74d_18", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 22.6}, {\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 26.6}]}", "target": "power strip", "split": "train"}
{"id": "5d591ff74d_19", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "5d591ff74d_20", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. trash can\nB. paper\nC. backpack\nD. box", "choices": ["A. trash can", "B. paper", "C. backpack", "D. box"], "answer": "B. paper", "target": "spray bottle", "split": "train"}
{"id": "5d591ff74d_21", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and above the paper, about 14 cm away. It is also to the side of and above the trash can, about 43 cm away.", "target": "spray bottle", "split": "train"}
{"id": "5d591ff74d_22", "scene_id": "5d591ff74d", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 13.5}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 43.3}]}", "target": "spray bottle", "split": "train"}
{"id": "5d902f1593_0", "scene_id": "5d902f1593", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "train"}
{"id": "5d902f1593_1", "scene_id": "5d902f1593", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cable duct\nB. power socket\nC. monitor\nD. table", "choices": ["A. cable duct", "B. power socket", "C. monitor", "D. table"], "answer": "B. power socket", "target": "backpack", "split": "train"}
{"id": "5d902f1593_2", "scene_id": "5d902f1593", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also next to the power socket, about 8 cm away.", "target": "backpack", "split": "train"}
{"id": "5d902f1593_3", "scene_id": "5d902f1593", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"power socket\", \"relation\": \"next_to\", \"distance_cm\": 8.5}]}", "target": "backpack", "split": "train"}
{"id": "5dd2b849be_0", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "5dd2b849be_1", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. remote\nB. chair\nC. webcam\nD. tv", "choices": ["A. remote", "B. chair", "C. webcam", "D. tv"], "answer": "A. remote", "target": "keyboard", "split": "train"}
{"id": "5dd2b849be_2", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the remote, about 9 cm away. It is also to the side of and below the tv, about 24 cm away. It is further near the chair, about 57 cm away.", "target": "keyboard", "split": "train"}
{"id": "5dd2b849be_3", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"tv stand\", \"references\": [{\"object\": \"remote\", \"relation\": \"next_to\", \"distance_cm\": 8.7}, {\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 24.4}]}", "target": "keyboard", "split": "train"}
{"id": "5dd2b849be_4", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "5dd2b849be_5", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. power strip\nB. chair\nC. keyboard\nD. tv", "choices": ["A. power strip", "B. chair", "C. keyboard", "D. tv"], "answer": "C. keyboard", "target": "remote", "split": "train"}
{"id": "5dd2b849be_6", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is next to the keyboard, about 9 cm away. It is also to the side of and below the tv, about 26 cm away. It is further to the side of and above the chair, about 52 cm away.", "target": "remote", "split": "train"}
{"id": "5dd2b849be_7", "scene_id": "5dd2b849be", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"tv stand\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.7}, {\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 26.4}]}", "target": "remote", "split": "train"}
{"id": "5eb31827b7_0", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "5eb31827b7_1", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mouse\nB. monitor\nC. trash can\nD. box", "choices": ["A. mouse", "B. monitor", "C. trash can", "D. box"], "answer": "B. monitor", "target": "bottle", "split": "train"}
{"id": "5eb31827b7_2", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the cabinet. It is also to the side of and below the monitor, about 34 cm away. It is further near the mouse, about 42 cm away.", "target": "bottle", "split": "train"}
{"id": "5eb31827b7_3", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 34.0}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 42.3}]}", "target": "bottle", "split": "train"}
{"id": "5eb31827b7_4", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "5eb31827b7_5", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headset in this scene?\nA. panel\nB. mouse\nC. monitor\nD. rolling chair", "choices": ["A. panel", "B. mouse", "C. monitor", "D. rolling chair"], "answer": "C. monitor", "target": "headset", "split": "train"}
{"id": "5eb31827b7_6", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is on the table. It is also under the monitor, about 4 cm away. It is further next to the mouse, about 15 cm away.", "target": "headset", "split": "train"}
{"id": "5eb31827b7_7", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 15.0}]}", "target": "headset", "split": "train"}
{"id": "5eb31827b7_8", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "5eb31827b7_9", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. paper\nC. laptop\nD. rolling chair", "choices": ["A. monitor", "B. paper", "C. laptop", "D. rolling chair"], "answer": "B. paper", "target": "keyboard", "split": "train"}
{"id": "5eb31827b7_10", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the paper, about 14 cm away. It is further to the side of and above the rolling chair, about 17 cm away.", "target": "keyboard", "split": "train"}
{"id": "5eb31827b7_11", "scene_id": "5eb31827b7", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 13.7}, {\"object\": \"rolling chair\", \"relation\": \"side_above\", \"distance_cm\": 16.6}]}", "target": "keyboard", "split": "train"}
{"id": "5fd8addedf_0", "scene_id": "5fd8addedf", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the comb appear?", "choices": null, "answer": "No", "target": "comb", "split": "train"}
{"id": "5fd8addedf_1", "scene_id": "5fd8addedf", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. chopping board\nB. paper towels\nC. kitchen rag\nD. paper", "choices": ["A. chopping board", "B. paper towels", "C. kitchen rag", "D. paper"], "answer": "D. paper", "target": "bottle", "split": "train"}
{"id": "5fd8addedf_2", "scene_id": "5fd8addedf", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the paper, about 10 cm away. It is further next to the paper towels, about 13 cm away.", "target": "bottle", "split": "train"}
{"id": "5fd8addedf_3", "scene_id": "5fd8addedf", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 10.3}, {\"object\": \"paper towels\", \"relation\": \"next_to\", \"distance_cm\": 12.8}]}", "target": "bottle", "split": "train"}
{"id": "603964346e_0", "scene_id": "603964346e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "603964346e_1", "scene_id": "603964346e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is to the side of and above the bottle, about 3 cm away. It is also to the side of and above the box, about 3 cm away. It is further to the side of and above the kitchen counter, about 4 cm away.", "target": "scissors", "split": "train"}
{"id": "603964346e_2", "scene_id": "603964346e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 3.2}]}", "target": "scissors", "split": "train"}
{"id": "6115eddb86_0", "scene_id": "6115eddb86", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "6115eddb86_1", "scene_id": "6115eddb86", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. box\nB. tv\nC. wall unit\nD. counter", "choices": ["A. box", "B. tv", "C. wall unit", "D. counter"], "answer": "C. wall unit", "target": "tv remote", "split": "train"}
{"id": "6115eddb86_2", "scene_id": "6115eddb86", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the cabinet. It is also to the side of and below the wall unit, about 17 cm away. It is further to the side of and above the counter, about 20 cm away.", "target": "tv remote", "split": "train"}
{"id": "6115eddb86_3", "scene_id": "6115eddb86", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"wall unit\", \"relation\": \"side_below\", \"distance_cm\": 17.3}, {\"object\": \"counter\", \"relation\": \"side_above\", \"distance_cm\": 19.9}]}", "target": "tv remote", "split": "train"}
{"id": "6183f0657d_0", "scene_id": "6183f0657d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "6183f0657d_1", "scene_id": "6183f0657d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. blanket\nB. cabinet\nC. book\nD. cup", "choices": ["A. blanket", "B. cabinet", "C. book", "D. cup"], "answer": "A. blanket", "target": "bottle", "split": "train"}
{"id": "6183f0657d_2", "scene_id": "6183f0657d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also to the side of and above the blanket, about 16 cm away. It is further near the book, about 31 cm away.", "target": "bottle", "split": "train"}
{"id": "6183f0657d_3", "scene_id": "6183f0657d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"blanket\", \"relation\": \"side_above\", \"distance_cm\": 16.2}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 30.9}]}", "target": "bottle", "split": "train"}
{"id": "61adeff7d5_0", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "61adeff7d5_1", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. sink\nB. knife\nC. jar\nD. bottle", "choices": ["A. sink", "B. knife", "C. jar", "D. bottle"], "answer": "C. jar", "target": "mug", "split": "train"}
{"id": "61adeff7d5_2", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is under the jar, about 4 cm away. It is also next to the knife, about 6 cm away. It is further next to the bottle, about 11 cm away.", "target": "mug", "split": "train"}
{"id": "61adeff7d5_3", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"jar\", \"relation\": \"under\", \"distance_cm\": 4.4}, {\"object\": \"knife\", \"relation\": \"next_to\", \"distance_cm\": 5.6}]}", "target": "mug", "split": "train"}
{"id": "61adeff7d5_4", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "61adeff7d5_5", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notepad in this scene?\nA. paper bag\nB. office chair\nC. backpack\nD. chair", "choices": ["A. paper bag", "B. office chair", "C. backpack", "D. chair"], "answer": "D. chair", "target": "notepad", "split": "train"}
{"id": "61adeff7d5_6", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notepad is on the table. It is also to the side of and above the chair, about 11 cm away. It is further to the side of and above the office chair, about 18 cm away.", "target": "notepad", "split": "train"}
{"id": "61adeff7d5_7", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"notepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 11.1}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 17.6}]}", "target": "notepad", "split": "train"}
{"id": "61adeff7d5_8", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "Yes", "target": "razor", "split": "train"}
{"id": "61adeff7d5_9", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the razor in this scene?\nA. bottle\nB. mirror\nC. towel\nD. pouch", "choices": ["A. bottle", "B. mirror", "C. towel", "D. pouch"], "answer": "B. mirror", "target": "razor", "split": "train"}
{"id": "61adeff7d5_10", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the razor in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The razor is on the rack. It is also next to the mirror. It is further to the side of and above the bottle, about 22 cm away.", "target": "razor", "split": "train"}
{"id": "61adeff7d5_11", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the razor in structured JSON format.", "choices": null, "answer": "{\"target\": \"razor\", \"support_surface\": \"rack\", \"references\": [{\"object\": \"mirror\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 22.3}]}", "target": "razor", "split": "train"}
{"id": "61adeff7d5_12", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "Yes", "target": "remote", "split": "train"}
{"id": "61adeff7d5_13", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. notebook\nB. box\nC. apples\nD. coffee table", "choices": ["A. notebook", "B. box", "C. apples", "D. coffee table"], "answer": "D. coffee table", "target": "remote", "split": "train"}
{"id": "61adeff7d5_14", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is to the side of and above the coffee table, about 6 cm away. It is also to the side of and above the notebook, about 8 cm away. It is further next to the box, about 8 cm away.", "target": "remote", "split": "train"}
{"id": "61adeff7d5_15", "scene_id": "61adeff7d5", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"coffee table\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"notebook\", \"relation\": \"side_above\", \"distance_cm\": 7.6}]}", "target": "remote", "split": "train"}
{"id": "623c973c56_0", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "623c973c56_1", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. fire extinguisher case\nB. bowl\nC. floodlight\nD. fire extinguisher", "choices": ["A. fire extinguisher case", "B. bowl", "C. floodlight", "D. fire extinguisher"], "answer": "A. fire extinguisher case", "target": "extension cord", "split": "train"}
{"id": "623c973c56_2", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the fire extinguisher case, about 48 cm away. It is also near the floodlight, about 58 cm away. It is further to the side of and below the fire extinguisher, about 59 cm away.", "target": "extension cord", "split": "train"}
{"id": "623c973c56_3", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"fire extinguisher case\", \"references\": [{\"object\": \"fire extinguisher case\", \"relation\": \"side_below\", \"distance_cm\": 47.6}, {\"object\": \"floodlight\", \"relation\": \"near\", \"distance_cm\": 57.8}]}", "target": "extension cord", "split": "train"}
{"id": "623c973c56_4", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "623c973c56_5", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. safety poster\nB. power outlet\nC. bowl\nD. glass cups", "choices": ["A. safety poster", "B. power outlet", "C. bowl", "D. glass cups"], "answer": "B. power outlet", "target": "phone charger", "split": "train"}
{"id": "623c973c56_6", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the glass cups, about 58 cm away. It is also to the side of and below the safety poster, about 60 cm away. It is further to the side of and below the bowl, about 79 cm away.", "target": "phone charger", "split": "train"}
{"id": "623c973c56_7", "scene_id": "623c973c56", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"power outlet\", \"references\": [{\"object\": \"glass cups\", \"relation\": \"side_below\", \"distance_cm\": 58.2}, {\"object\": \"safety poster\", \"relation\": \"side_below\", \"distance_cm\": 59.7}]}", "target": "phone charger", "split": "train"}
{"id": "635852d56e_0", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "635852d56e_1", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. box\nB. flower pot\nC. books\nD. shelf", "choices": ["A. box", "B. flower pot", "C. books", "D. shelf"], "answer": "B. flower pot", "target": "bottle", "split": "train"}
{"id": "635852d56e_2", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is next to the flower pot, about 6 cm away. It is also to the side of and above the books, about 10 cm away. It is further to the side of and above the shelf, about 39 cm away.", "target": "bottle", "split": "train"}
{"id": "635852d56e_3", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"flower pot\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 10.1}]}", "target": "bottle", "split": "train"}
{"id": "635852d56e_4", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "635852d56e_5", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. bowl\nB. shelf\nC. water jug\nD. bottle", "choices": ["A. bowl", "B. shelf", "C. water jug", "D. bottle"], "answer": "B. shelf", "target": "cup", "split": "train"}
{"id": "635852d56e_6", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the shelf. It is also near the bottle, about 33 cm away. It is further to the side of and below the water jug, about 36 cm away.", "target": "cup", "split": "train"}
{"id": "635852d56e_7", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bottle\", \"relation\": \"near\", \"distance_cm\": 32.9}, {\"object\": \"water jug\", \"relation\": \"side_below\", \"distance_cm\": 35.6}]}", "target": "cup", "split": "train"}
{"id": "635852d56e_8", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "635852d56e_9", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. chair\nB. jacket\nC. monitor\nD. desk", "choices": ["A. chair", "B. jacket", "C. monitor", "D. desk"], "answer": "D. desk", "target": "power adapter", "split": "train"}
{"id": "635852d56e_10", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and above the desk, about 3 cm away. It is also to the side of and below the monitor, about 14 cm away. It is further to the side of and above the chair, about 27 cm away.", "target": "power adapter", "split": "train"}
{"id": "635852d56e_11", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 3.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.9}]}", "target": "power adapter", "split": "train"}
{"id": "635852d56e_12", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "635852d56e_13", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor\nB. laptop\nC. jacket\nD. chair", "choices": ["A. monitor", "B. laptop", "C. jacket", "D. chair"], "answer": "A. monitor", "target": "power strip", "split": "train"}
{"id": "635852d56e_14", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the desk. It is also to the side of and below the monitor, about 10 cm away. It is further to the side of and above the chair, about 26 cm away.", "target": "power strip", "split": "train"}
{"id": "635852d56e_15", "scene_id": "635852d56e", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 26.2}]}", "target": "power strip", "split": "train"}
{"id": "6391f0f921_0", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "6391f0f921_1", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. laptop holder\nB. laptop\nC. paper\nD. heater", "choices": ["A. laptop holder", "B. laptop", "C. paper", "D. heater"], "answer": "C. paper", "target": "keyboard", "split": "train"}
{"id": "6391f0f921_2", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the paper, about 9 cm away. It is further near the laptop holder, about 16 cm away.", "target": "keyboard", "split": "train"}
{"id": "6391f0f921_3", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 8.5}, {\"object\": \"laptop holder\", \"relation\": \"near\", \"distance_cm\": 15.8}]}", "target": "keyboard", "split": "train"}
{"id": "6391f0f921_4", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "6391f0f921_5", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. desk\nB. office chair\nC. laptop holder\nD. laptop", "choices": ["A. desk", "B. office chair", "C. laptop holder", "D. laptop"], "answer": "A. desk", "target": "mouse", "split": "train"}
{"id": "6391f0f921_6", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the desk, about 5 cm away. It is also to the side of and below the laptop, about 11 cm away. It is further to the side of and above the office chair, about 16 cm away.", "target": "mouse", "split": "train"}
{"id": "6391f0f921_7", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 4.8}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 10.6}]}", "target": "mouse", "split": "train"}
{"id": "6391f0f921_8", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "6391f0f921_9", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. office chair\nB. paper\nC. laptop holder\nD. laptop", "choices": ["A. office chair", "B. paper", "C. laptop holder", "D. laptop"], "answer": "A. office chair", "target": "mousepad", "split": "train"}
{"id": "6391f0f921_10", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the desk. It is also to the side of and above the office chair, about 7 cm away. It is further to the side of and below the laptop, about 8 cm away.", "target": "mousepad", "split": "train"}
{"id": "6391f0f921_11", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 6.8}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 8.4}]}", "target": "mousepad", "split": "train"}
{"id": "6391f0f921_12", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "6391f0f921_13", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the clothes, about 26 cm away. It is also near the teapot, about 27 cm away. It is further to the side of and below the rice cooker, about 29 cm away.", "target": "phone charger", "split": "train"}
{"id": "6391f0f921_14", "scene_id": "6391f0f921", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"desk fan\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_below\", \"distance_cm\": 26.5}, {\"object\": \"teapot\", \"relation\": \"near\", \"distance_cm\": 26.8}]}", "target": "phone charger", "split": "train"}
{"id": "652d9cb0d7_0", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "No", "target": "handbag", "split": "train"}
{"id": "652d9cb0d7_1", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "652d9cb0d7_2", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. heater\nB. windowsill\nC. office chair\nD. power strip", "choices": ["A. heater", "B. windowsill", "C. office chair", "D. power strip"], "answer": "C. office chair", "target": "keyboard", "split": "train"}
{"id": "652d9cb0d7_3", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the office chair, about 23 cm away. It is further near the power strip, about 27 cm away.", "target": "keyboard", "split": "train"}
{"id": "652d9cb0d7_4", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 22.9}, {\"object\": \"power strip\", \"relation\": \"near\", \"distance_cm\": 26.7}]}", "target": "keyboard", "split": "train"}
{"id": "652d9cb0d7_5", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "652d9cb0d7_6", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil case in this scene?\nA. cabinet\nB. backpack\nC. curtain\nD. cactus", "choices": ["A. cabinet", "B. backpack", "C. curtain", "D. cactus"], "answer": "A. cabinet", "target": "pencil case", "split": "train"}
{"id": "652d9cb0d7_7", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil case is to the side of and above the cabinet, about 3 cm away. It is also to the side of and below the curtain, about 18 cm away.", "target": "pencil case", "split": "train"}
{"id": "652d9cb0d7_8", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 18.2}]}", "target": "pencil case", "split": "train"}
{"id": "652d9cb0d7_9", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "652d9cb0d7_10", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. lan outlet\nB. windowsill\nC. coax outlet\nD. plant pot", "choices": ["A. lan outlet", "B. windowsill", "C. coax outlet", "D. plant pot"], "answer": "A. lan outlet", "target": "phone charger", "split": "train"}
{"id": "652d9cb0d7_11", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is above the lan outlet, about 19 cm away. It is also to the side of and below the windowsill, about 22 cm away. It is further to the side of and below the plant pot, about 28 cm away.", "target": "phone charger", "split": "train"}
{"id": "652d9cb0d7_12", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"table lamp\", \"references\": [{\"object\": \"lan outlet\", \"relation\": \"above\", \"distance_cm\": 19.2}, {\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 22.2}]}", "target": "phone charger", "split": "train"}
{"id": "652d9cb0d7_13", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "652d9cb0d7_14", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. windowsill\nB. keyboard\nC. monitor\nD. window frame", "choices": ["A. windowsill", "B. keyboard", "C. monitor", "D. window frame"], "answer": "A. windowsill", "target": "power strip", "split": "train"}
{"id": "652d9cb0d7_15", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also next to the windowsill, about 7 cm away. It is further to the side of and below the monitor, about 17 cm away.", "target": "power strip", "split": "train"}
{"id": "652d9cb0d7_16", "scene_id": "652d9cb0d7", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"windowsill\", \"relation\": \"next_to\", \"distance_cm\": 7.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 16.9}]}", "target": "power strip", "split": "train"}
{"id": "654a4f341b_0", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "Yes", "target": "adapter", "split": "train"}
{"id": "654a4f341b_1", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the adapter in this scene?\nA. mouse\nB. laptop\nC. mousepad\nD. water filter", "choices": ["A. mouse", "B. laptop", "C. mousepad", "D. water filter"], "answer": "B. laptop", "target": "adapter", "split": "train"}
{"id": "654a4f341b_2", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The adapter is on the table. It is also to the side of and below the laptop, about 16 cm away. It is further near the mousepad, about 23 cm away.", "target": "adapter", "split": "train"}
{"id": "654a4f341b_3", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 15.9}, {\"object\": \"mousepad\", \"relation\": \"near\", \"distance_cm\": 23.0}]}", "target": "adapter", "split": "train"}
{"id": "654a4f341b_4", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "654a4f341b_5", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. paper\nB. chair\nC. mousepad\nD. water filter", "choices": ["A. paper", "B. chair", "C. mousepad", "D. water filter"], "answer": "B. chair", "target": "backpack", "split": "train"}
{"id": "654a4f341b_6", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 36 cm away. It is also to the side of and below the paper, about 56 cm away. It is further to the side of and below the water filter, about 56 cm away.", "target": "backpack", "split": "train"}
{"id": "654a4f341b_7", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 36.4}, {\"object\": \"paper\", \"relation\": \"side_below\", \"distance_cm\": 55.9}]}", "target": "backpack", "split": "train"}
{"id": "654a4f341b_8", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "654a4f341b_9", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. mouse\nC. chair\nD. adapter", "choices": ["A. monitor", "B. mouse", "C. chair", "D. adapter"], "answer": "A. monitor", "target": "keyboard", "split": "train"}
{"id": "654a4f341b_10", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the monitor, about 12 cm away. It is further to the side of and above the chair, about 20 cm away.", "target": "keyboard", "split": "train"}
{"id": "654a4f341b_11", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 20.5}]}", "target": "keyboard", "split": "train"}
{"id": "654a4f341b_12", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "No", "target": "pen", "split": "train"}
{"id": "654a4f341b_13", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. chair\nB. water filter\nC. monitor\nD. laptop", "choices": ["A. chair", "B. water filter", "C. monitor", "D. laptop"], "answer": "D. laptop", "target": "mousepad", "split": "train"}
{"id": "654a4f341b_14", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also under the laptop, about 3 cm away. It is further to the side of and above the chair, about 5 cm away.", "target": "mousepad", "split": "train"}
{"id": "654a4f341b_15", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"under\", \"distance_cm\": 2.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 4.7}]}", "target": "mousepad", "split": "train"}
{"id": "654a4f341b_16", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "654a4f341b_17", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. soap bottle\nB. bottle\nC. shower wall\nD. plant", "choices": ["A. soap bottle", "B. bottle", "C. shower wall", "D. plant"], "answer": "A. soap bottle", "target": "mug", "split": "train"}
{"id": "654a4f341b_18", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is under the soap bottle. It is also to the side of and below the plant, about 11 cm away. It is further to the side of and below the shower wall, about 15 cm away.", "target": "mug", "split": "train"}
{"id": "654a4f341b_19", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"soap bottle\", \"relation\": \"under\", \"distance_cm\": 2.5}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 10.6}]}", "target": "mug", "split": "train"}
{"id": "654a4f341b_20", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "654a4f341b_21", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. cabinet\nB. pedestal fan\nC. light switch\nD. door frame", "choices": ["A. cabinet", "B. pedestal fan", "C. light switch", "D. door frame"], "answer": "D. door frame", "target": "power adapter", "split": "train"}
{"id": "654a4f341b_22", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and below the door frame, about 10 cm away. It is also to the side of and below the light switch, about 17 cm away. It is further to the side of and below the pedestal fan, about 20 cm away.", "target": "power adapter", "split": "train"}
{"id": "654a4f341b_23", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"door frame\", \"relation\": \"side_below\", \"distance_cm\": 10.1}, {\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 16.6}]}", "target": "power adapter", "split": "train"}
{"id": "654a4f341b_24", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "654a4f341b_25", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power plug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power plug is to the side of and above the rug, about 17 cm away. It is also to the side of and below the door, about 22 cm away.", "target": "power plug", "split": "train"}
{"id": "654a4f341b_26", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power plug in structured JSON format.", "choices": null, "answer": "{\"target\": \"power plug\", \"support_surface\": \"rug\", \"references\": [{\"object\": \"rug\", \"relation\": \"side_above\", \"distance_cm\": 16.7}, {\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 21.7}]}", "target": "power plug", "split": "train"}
{"id": "654a4f341b_27", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "Yes", "target": "toothbrush", "split": "train"}
{"id": "654a4f341b_28", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is on the table. It is also to the side of and above the book, about 8 cm away. It is further to the side of and above the laundry basket, about 9 cm away.", "target": "toothbrush", "split": "train"}
{"id": "654a4f341b_29", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 8.4}, {\"object\": \"laundry basket\", \"relation\": \"side_above\", \"distance_cm\": 8.8}]}", "target": "toothbrush", "split": "train"}
{"id": "654a4f341b_30", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "654a4f341b_31", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. air fryer\nB. adapter\nC. box\nD. bag", "choices": ["A. air fryer", "B. adapter", "C. box", "D. bag"], "answer": "C. box", "target": "tote bag", "split": "train"}
{"id": "654a4f341b_32", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and below the box, about 19 cm away. It is also to the side of and below the bag, about 46 cm away.", "target": "tote bag", "split": "train"}
{"id": "654a4f341b_33", "scene_id": "654a4f341b", "source_dataset": "scannetpp", "local_sample_id": "33", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"box\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 19.5}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 46.0}]}", "target": "tote bag", "split": "train"}
{"id": "65aa268fcb_0", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "65aa268fcb_1", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. power socket\nC. jacket\nD. books", "choices": ["A. chair", "B. power socket", "C. jacket", "D. books"], "answer": "B. power socket", "target": "backpack", "split": "train"}
{"id": "65aa268fcb_2", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the power socket, about 12 cm away. It is also to the side of and below the jacket, about 16 cm away. It is further to the side of and below the chair, about 32 cm away.", "target": "backpack", "split": "train"}
{"id": "65aa268fcb_3", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 11.7}, {\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 15.9}]}", "target": "backpack", "split": "train"}
{"id": "65aa268fcb_4", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "65aa268fcb_5", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. table cloth\nB. stool\nC. lamp\nD. bottle", "choices": ["A. table cloth", "B. stool", "C. lamp", "D. bottle"], "answer": "A. table cloth", "target": "cup", "split": "train"}
{"id": "65aa268fcb_6", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the table cloth, about 7 cm away. It is also to the side of and below the bottle, about 16 cm away. It is further to the side of and above the stool, about 55 cm away.", "target": "cup", "split": "train"}
{"id": "65aa268fcb_7", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table cloth\", \"references\": [{\"object\": \"table cloth\", \"relation\": \"next_to\", \"distance_cm\": 6.7}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 16.4}]}", "target": "cup", "split": "train"}
{"id": "65aa268fcb_8", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "65aa268fcb_9", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. desk calendar\nB. laptop\nC. dustbin\nD. jacket", "choices": ["A. desk calendar", "B. laptop", "C. dustbin", "D. jacket"], "answer": "A. desk calendar", "target": "headphones", "split": "train"}
{"id": "65aa268fcb_10", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also to the side of and below the desk calendar, about 6 cm away. It is further to the side of and above the jacket, about 35 cm away.", "target": "headphones", "split": "train"}
{"id": "65aa268fcb_11", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"desk calendar\", \"relation\": \"side_below\", \"distance_cm\": 6.0}, {\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 34.6}]}", "target": "headphones", "split": "train"}
{"id": "65aa268fcb_12", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "65aa268fcb_13", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. laptop\nB. water jug\nC. curtain\nD. bottle", "choices": ["A. laptop", "B. water jug", "C. curtain", "D. bottle"], "answer": "B. water jug", "target": "mug", "split": "train"}
{"id": "65aa268fcb_14", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and below the water jug, about 7 cm away. It is further to the side of and below the curtain, about 15 cm away.", "target": "mug", "split": "train"}
{"id": "65aa268fcb_15", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"water jug\", \"relation\": \"side_below\", \"distance_cm\": 7.4}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 14.7}]}", "target": "mug", "split": "train"}
{"id": "65aa268fcb_16", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "65aa268fcb_17", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. stool\nB. dustbin\nC. refrigerator\nD. chair", "choices": ["A. stool", "B. dustbin", "C. refrigerator", "D. chair"], "answer": "A. stool", "target": "power strip", "split": "train"}
{"id": "65aa268fcb_18", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the stool. It is also to the side of and below the chair, about 14 cm away. It is further to the side of and below the refrigerator, about 54 cm away.", "target": "power strip", "split": "train"}
{"id": "65aa268fcb_19", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"stool\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 14.0}]}", "target": "power strip", "split": "train"}
{"id": "65aa268fcb_20", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "65aa268fcb_21", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. laptop\nB. bottle\nC. curtain\nD. water jug", "choices": ["A. laptop", "B. bottle", "C. curtain", "D. water jug"], "answer": "C. curtain", "target": "tissue box", "split": "train"}
{"id": "65aa268fcb_22", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also under the curtain. It is further under the water jug, about 3 cm away.", "target": "tissue box", "split": "train"}
{"id": "65aa268fcb_23", "scene_id": "65aa268fcb", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"curtain\", \"relation\": \"under\", \"distance_cm\": 2.3}, {\"object\": \"water jug\", \"relation\": \"under\", \"distance_cm\": 3.4}]}", "target": "tissue box", "split": "train"}
{"id": "66ba53719a_0", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "66ba53719a_1", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. plant\nB. books\nC. box\nD. coaster", "choices": ["A. plant", "B. books", "C. box", "D. coaster"], "answer": "D. coaster", "target": "bottle", "split": "train"}
{"id": "66ba53719a_2", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the coaster, about 14 cm away. It is also to the side of and below the plant, about 34 cm away. It is further to the side of and above the books, about 36 cm away.", "target": "bottle", "split": "train"}
{"id": "66ba53719a_3", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"carpet\", \"references\": [{\"object\": \"coaster\", \"relation\": \"side_above\", \"distance_cm\": 14.4}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 34.2}]}", "target": "bottle", "split": "train"}
{"id": "66ba53719a_4", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "Yes", "target": "game controller", "split": "train"}
{"id": "66ba53719a_5", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the game controller in this scene?\nA. cabinet\nB. computer tower\nC. box\nD. whiteboard", "choices": ["A. cabinet", "B. computer tower", "C. box", "D. whiteboard"], "answer": "B. computer tower", "target": "game controller", "split": "train"}
{"id": "66ba53719a_6", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the game controller in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The game controller is to the side of and above the computer tower, about 7 cm away. It is also to the side of and above the whiteboard, about 10 cm away. It is further to the side of and below the cabinet, about 14 cm away.", "target": "game controller", "split": "train"}
{"id": "66ba53719a_7", "scene_id": "66ba53719a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the game controller in structured JSON format.", "choices": null, "answer": "{\"target\": \"game controller\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 7.3}, {\"object\": \"whiteboard\", \"relation\": \"side_above\", \"distance_cm\": 9.9}]}", "target": "game controller", "split": "train"}
{"id": "67d702f2e8_0", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "67d702f2e8_1", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. paper\nB. mug\nC. pens\nD. file rack", "choices": ["A. paper", "B. mug", "C. pens", "D. file rack"], "answer": "C. pens", "target": "bottle", "split": "train"}
{"id": "67d702f2e8_2", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the pens, about 4 cm away. It is further next to the mug, about 9 cm away.", "target": "bottle", "split": "train"}
{"id": "67d702f2e8_3", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pens\", \"relation\": \"side_above\", \"distance_cm\": 4.5}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 8.6}]}", "target": "bottle", "split": "train"}
{"id": "67d702f2e8_4", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "No", "target": "notebook", "split": "train"}
{"id": "67d702f2e8_5", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headset in this scene?\nA. monitor\nB. book\nC. keyboard\nD. paper", "choices": ["A. monitor", "B. book", "C. keyboard", "D. paper"], "answer": "D. paper", "target": "headset", "split": "train"}
{"id": "67d702f2e8_6", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is on the table. It is also next to the paper, about 3 cm away. It is further under the monitor, about 4 cm away.", "target": "headset", "split": "train"}
{"id": "67d702f2e8_7", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.5}]}", "target": "headset", "split": "train"}
{"id": "67d702f2e8_8", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "67d702f2e8_9", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. books\nB. paper\nC. bottle\nD. file rack", "choices": ["A. books", "B. paper", "C. bottle", "D. file rack"], "answer": "D. file rack", "target": "mug", "split": "train"}
{"id": "67d702f2e8_10", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and below the file rack, about 7 cm away. It is further next to the bottle, about 9 cm away.", "target": "mug", "split": "train"}
{"id": "67d702f2e8_11", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"file rack\", \"relation\": \"side_below\", \"distance_cm\": 6.6}, {\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 8.6}]}", "target": "mug", "split": "train"}
{"id": "67d702f2e8_12", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "No", "target": "soap", "split": "train"}
{"id": "67d702f2e8_13", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. telephone\nB. table lamp\nC. keyboard\nD. monitor", "choices": ["A. telephone", "B. table lamp", "C. keyboard", "D. monitor"], "answer": "B. table lamp", "target": "pen holder", "split": "train"}
{"id": "67d702f2e8_14", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also under the table lamp, about 5 cm away. It is further next to the telephone, about 8 cm away.", "target": "pen holder", "split": "train"}
{"id": "67d702f2e8_15", "scene_id": "67d702f2e8", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"under\", \"distance_cm\": 4.6}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 8.0}]}", "target": "pen holder", "split": "train"}
{"id": "684d4da1fe_0", "scene_id": "684d4da1fe", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "684d4da1fe_1", "scene_id": "684d4da1fe", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. sink\nB. soap bottle\nC. power socket\nD. faucet", "choices": ["A. sink", "B. soap bottle", "C. power socket", "D. faucet"], "answer": "D. faucet", "target": "toothbrush", "split": "train"}
{"id": "684d4da1fe_2", "scene_id": "684d4da1fe", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is to the side of and above the faucet, about 15 cm away. It is also to the side of and above the sink, about 21 cm away. It is further to the side of and above the soap bottle, about 26 cm away.", "target": "toothbrush", "split": "train"}
{"id": "684d4da1fe_3", "scene_id": "684d4da1fe", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"bathroom shelf\", \"references\": [{\"object\": \"faucet\", \"relation\": \"side_above\", \"distance_cm\": 14.7}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 21.1}]}", "target": "toothbrush", "split": "train"}
{"id": "698c3bf2bb_0", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "698c3bf2bb_1", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. heater\nB. window frame\nC. curtain\nD. desk", "choices": ["A. heater", "B. window frame", "C. curtain", "D. desk"], "answer": "D. desk", "target": "backpack", "split": "train"}
{"id": "698c3bf2bb_2", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the desk, about 8 cm away. It is also to the side of and above the heater, about 17 cm away. It is further to the side of and below the curtain, about 20 cm away.", "target": "backpack", "split": "train"}
{"id": "698c3bf2bb_3", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 8.0}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 16.7}]}", "target": "backpack", "split": "train"}
{"id": "698c3bf2bb_4", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "698c3bf2bb_5", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. floor lamp\nB. portable heater\nC. coax outlet\nD. office chair", "choices": ["A. floor lamp", "B. portable heater", "C. coax outlet", "D. office chair"], "answer": "A. floor lamp", "target": "extension cord", "split": "train"}
{"id": "698c3bf2bb_6", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the desk. It is also to the side of and below the floor lamp, about 10 cm away. It is further to the side of and above the portable heater, about 19 cm away.", "target": "extension cord", "split": "train"}
{"id": "698c3bf2bb_7", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"floor lamp\", \"relation\": \"side_below\", \"distance_cm\": 9.5}, {\"object\": \"portable heater\", \"relation\": \"side_above\", \"distance_cm\": 18.5}]}", "target": "extension cord", "split": "train"}
{"id": "698c3bf2bb_8", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "698c3bf2bb_9", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. pillow\nB. shelf rail\nC. bowl\nD. bottle", "choices": ["A. pillow", "B. shelf rail", "C. bowl", "D. bottle"], "answer": "A. pillow", "target": "phone charger", "split": "train"}
{"id": "698c3bf2bb_10", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and above the pillow, about 20 cm away. It is also to the side of and below the shelf rail, about 33 cm away. It is further to the side of and below the bowl, about 54 cm away.", "target": "phone charger", "split": "train"}
{"id": "698c3bf2bb_11", "scene_id": "698c3bf2bb", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"shelf rail\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 19.9}, {\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 32.6}]}", "target": "phone charger", "split": "train"}
{"id": "69e56cf0f8_0", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "69e56cf0f8_1", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also to the side of and above the basket, about 5 cm away. It is further to the side of and above the desk lamp, about 25 cm away.", "target": "bottle", "split": "train"}
{"id": "69e56cf0f8_2", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"basket\", \"relation\": \"side_above\", \"distance_cm\": 4.5}, {\"object\": \"desk lamp\", \"relation\": \"side_above\", \"distance_cm\": 25.5}]}", "target": "bottle", "split": "train"}
{"id": "69e56cf0f8_3", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "69e56cf0f8_4", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. chair\nB. monitor\nC. router\nD. table", "choices": ["A. chair", "B. monitor", "C. router", "D. table"], "answer": "A. chair", "target": "notebook", "split": "train"}
{"id": "69e56cf0f8_5", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and above the chair, about 13 cm away. It is also to the side of and above the table, about 15 cm away. It is further to the side of and below the monitor, about 19 cm away.", "target": "notebook", "split": "train"}
{"id": "69e56cf0f8_6", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table cloth\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 13.3}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 14.5}]}", "target": "notebook", "split": "train"}
{"id": "69e56cf0f8_7", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "69e56cf0f8_8", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. cup\nB. candle\nC. plant\nD. ball", "choices": ["A. cup", "B. candle", "C. plant", "D. ball"], "answer": "C. plant", "target": "water bottle", "split": "train"}
{"id": "69e56cf0f8_9", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is under the plant, about 4 cm away. It is also to the side of and above the ball, about 15 cm away. It is further to the side of and above the candle, about 20 cm away.", "target": "water bottle", "split": "train"}
{"id": "69e56cf0f8_10", "scene_id": "69e56cf0f8", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table cloth\", \"references\": [{\"object\": \"plant\", \"relation\": \"under\", \"distance_cm\": 4.2}, {\"object\": \"ball\", \"relation\": \"side_above\", \"distance_cm\": 14.7}]}", "target": "water bottle", "split": "train"}
{"id": "69e5939669_0", "scene_id": "69e5939669", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "69e5939669_1", "scene_id": "69e5939669", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. paper\nB. monitor\nC. table\nD. stool", "choices": ["A. paper", "B. monitor", "C. table", "D. stool"], "answer": "A. paper", "target": "pen holder", "split": "train"}
{"id": "69e5939669_2", "scene_id": "69e5939669", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also to the side of and above the paper, about 11 cm away. It is further to the side of and above the stool, about 43 cm away.", "target": "pen holder", "split": "train"}
{"id": "69e5939669_3", "scene_id": "69e5939669", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 10.7}, {\"object\": \"stool\", \"relation\": \"side_above\", \"distance_cm\": 43.3}]}", "target": "pen holder", "split": "train"}
{"id": "6a217fbcb0_0", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "6a217fbcb0_1", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. window frame\nB. bowl\nC. toilet\nD. toilet paper holder", "choices": ["A. window frame", "B. bowl", "C. toilet", "D. toilet paper holder"], "answer": "A. window frame", "target": "bottle", "split": "train"}
{"id": "6a217fbcb0_2", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is attached to the window. It is also under the window frame, about 4 cm away. It is further to the side of and above the toilet paper holder, about 25 cm away.", "target": "bottle", "split": "train"}
{"id": "6a217fbcb0_3", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"window\", \"references\": [{\"object\": \"window frame\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"toilet paper holder\", \"relation\": \"side_above\", \"distance_cm\": 25.1}]}", "target": "bottle", "split": "train"}
{"id": "6a217fbcb0_4", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "6a217fbcb0_5", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. decoration\nB. cabinet\nC. hand soap bottle\nD. power sockets", "choices": ["A. decoration", "B. cabinet", "C. hand soap bottle", "D. power sockets"], "answer": "B. cabinet", "target": "cup", "split": "train"}
{"id": "6a217fbcb0_6", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the cabinet, about 27 cm away. It is also near the hand soap bottle, about 28 cm away. It is further to the side of and below the decoration, about 44 cm away.", "target": "cup", "split": "train"}
{"id": "6a217fbcb0_7", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 26.6}, {\"object\": \"hand soap bottle\", \"relation\": \"near\", \"distance_cm\": 27.7}]}", "target": "cup", "split": "train"}
{"id": "6a217fbcb0_8", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "6a217fbcb0_9", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. bowl\nB. plant\nC. plant pot\nD. power sockets", "choices": ["A. bowl", "B. plant", "C. plant pot", "D. power sockets"], "answer": "B. plant", "target": "tissue box", "split": "train"}
{"id": "6a217fbcb0_10", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and above the plant, about 19 cm away. It is also near the bowl, about 22 cm away. It is further to the side of and above the plant pot, about 22 cm away.", "target": "tissue box", "split": "train"}
{"id": "6a217fbcb0_11", "scene_id": "6a217fbcb0", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"plant\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_above\", \"distance_cm\": 18.8}, {\"object\": \"bowl\", \"relation\": \"near\", \"distance_cm\": 21.7}]}", "target": "tissue box", "split": "train"}
{"id": "6ad6cef000_0", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "Yes", "target": "eyeglasses", "split": "train"}
{"id": "6ad6cef000_1", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the eyeglasses in this scene?\nA. phone charger\nB. bedside lamp\nC. pillow\nD. power sockets", "choices": ["A. phone charger", "B. bedside lamp", "C. pillow", "D. power sockets"], "answer": "C. pillow", "target": "eyeglasses", "split": "train"}
{"id": "6ad6cef000_2", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the eyeglasses in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The eyeglasses is next to the pillow, about 12 cm away. It is also to the side of and below the bedside lamp, about 15 cm away. It is further to the side of and above the power sockets, about 39 cm away.", "target": "eyeglasses", "split": "train"}
{"id": "6ad6cef000_3", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the eyeglasses in structured JSON format.", "choices": null, "answer": "{\"target\": \"eyeglasses\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"pillow\", \"relation\": \"next_to\", \"distance_cm\": 11.9}, {\"object\": \"bedside lamp\", \"relation\": \"side_below\", \"distance_cm\": 14.5}]}", "target": "eyeglasses", "split": "train"}
{"id": "6ad6cef000_4", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "6ad6cef000_5", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. drawer cabinet\nB. monitor\nC. mouse\nD. office chair", "choices": ["A. drawer cabinet", "B. monitor", "C. mouse", "D. office chair"], "answer": "D. office chair", "target": "keyboard", "split": "train"}
{"id": "6ad6cef000_6", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and above the office chair, about 4 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "keyboard", "split": "train"}
{"id": "6ad6cef000_7", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 3.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.9}]}", "target": "keyboard", "split": "train"}
{"id": "6ad6cef000_8", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "6ad6cef000_9", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. drawer cabinet\nB. pc tower\nC. cabinet\nD. keyboard", "choices": ["A. drawer cabinet", "B. pc tower", "C. cabinet", "D. keyboard"], "answer": "A. drawer cabinet", "target": "mouse", "split": "train"}
{"id": "6ad6cef000_10", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also to the side of and above the drawer cabinet, about 10 cm away. It is further near the keyboard, about 16 cm away.", "target": "mouse", "split": "train"}
{"id": "6ad6cef000_11", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"drawer cabinet\", \"relation\": \"side_above\", \"distance_cm\": 10.1}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.8}]}", "target": "mouse", "split": "train"}
{"id": "6ad6cef000_12", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "6ad6cef000_13", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. phone charger\nB. pillow\nC. bedside lamp\nD. power sockets", "choices": ["A. phone charger", "B. pillow", "C. bedside lamp", "D. power sockets"], "answer": "B. pillow", "target": "notebook", "split": "train"}
{"id": "6ad6cef000_14", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is next to the pillow, about 11 cm away. It is also to the side of and below the bedside lamp, about 14 cm away. It is further to the side of and above the power sockets, about 18 cm away.", "target": "notebook", "split": "train"}
{"id": "6ad6cef000_15", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"pillow\", \"relation\": \"next_to\", \"distance_cm\": 10.6}, {\"object\": \"bedside lamp\", \"relation\": \"side_below\", \"distance_cm\": 14.2}]}", "target": "notebook", "split": "train"}
{"id": "6ad6cef000_16", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "6ad6cef000_17", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the door, about 21 cm away.", "target": "phone charger", "split": "train"}
{"id": "6ad6cef000_18", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 20.6}]}", "target": "phone charger", "split": "train"}
{"id": "6ad6cef000_19", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "6ad6cef000_20", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. desk\nB. pc tower\nC. office chair\nD. drawer cabinet", "choices": ["A. desk", "B. pc tower", "C. office chair", "D. drawer cabinet"], "answer": "A. desk", "target": "power strip", "split": "train"}
{"id": "6ad6cef000_21", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the desk, about 4 cm away. It is also to the side of and below the pc tower, about 12 cm away. It is further to the side of and below the office chair, about 41 cm away.", "target": "power strip", "split": "train"}
{"id": "6ad6cef000_22", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"pc tower\", \"relation\": \"side_below\", \"distance_cm\": 12.0}]}", "target": "power strip", "split": "train"}
{"id": "6ad6cef000_23", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "6ad6cef000_24", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. trash can\nB. toilet\nC. toilet paper holder\nD. toilet brush", "choices": ["A. trash can", "B. toilet", "C. toilet paper holder", "D. toilet brush"], "answer": "A. trash can", "target": "spray bottle", "split": "train"}
{"id": "6ad6cef000_25", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and below the trash can, about 9 cm away. It is also next to the toilet brush, about 14 cm away. It is further to the side of and below the toilet, about 22 cm away.", "target": "spray bottle", "split": "train"}
{"id": "6ad6cef000_26", "scene_id": "6ad6cef000", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"toilet paper holder\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 9.2}, {\"object\": \"toilet brush\", \"relation\": \"next_to\", \"distance_cm\": 14.1}]}", "target": "spray bottle", "split": "train"}
{"id": "6b40d1a939_0", "scene_id": "6b40d1a939", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "6b40d1a939_1", "scene_id": "6b40d1a939", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. bedpost\nB. cabinet\nC. water bottle\nD. bedside lamp", "choices": ["A. bedpost", "B. cabinet", "C. water bottle", "D. bedside lamp"], "answer": "A. bedpost", "target": "tissue box", "split": "train"}
{"id": "6b40d1a939_2", "scene_id": "6b40d1a939", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is under the bedpost, about 4 cm away. It is also to the side of and below the bedside lamp, about 9 cm away. It is further next to the water bottle, about 10 cm away.", "target": "tissue box", "split": "train"}
{"id": "6b40d1a939_3", "scene_id": "6b40d1a939", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"bedpost\", \"relation\": \"under\", \"distance_cm\": 3.9}, {\"object\": \"bedside lamp\", \"relation\": \"side_below\", \"distance_cm\": 8.6}]}", "target": "tissue box", "split": "train"}
{"id": "6b40d1a939_4", "scene_id": "6b40d1a939", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "train"}
{"id": "6b40d1a939_5", "scene_id": "6b40d1a939", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the cabinet. It is also next to the tissue box, about 10 cm away. It is further next to the tissue paper, about 10 cm away.", "target": "water bottle", "split": "train"}
{"id": "6b40d1a939_6", "scene_id": "6b40d1a939", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"tissue box\", \"relation\": \"next_to\", \"distance_cm\": 9.9}, {\"object\": \"tissue paper\", \"relation\": \"next_to\", \"distance_cm\": 9.9}]}", "target": "water bottle", "split": "train"}
{"id": "6bd39ac392_0", "scene_id": "6bd39ac392", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "train"}
{"id": "6bd39ac392_1", "scene_id": "6bd39ac392", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the cabinet.", "target": "pen holder", "split": "train"}
{"id": "6bd39ac392_2", "scene_id": "6bd39ac392", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"cabinet\", \"references\": []}", "target": "pen holder", "split": "train"}
{"id": "6d89a7320d_0", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "Yes", "target": "mouse pad", "split": "train"}
{"id": "6d89a7320d_1", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. monitor\nB. office chair\nC. power strip\nD. window sill", "choices": ["A. monitor", "B. office chair", "C. power strip", "D. window sill"], "answer": "B. office chair", "target": "mouse pad", "split": "train"}
{"id": "6d89a7320d_2", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the table. It is also to the side of and above the office chair, about 16 cm away. It is further to the side of and below the monitor, about 28 cm away.", "target": "mouse pad", "split": "train"}
{"id": "6d89a7320d_3", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 15.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 27.8}]}", "target": "mouse pad", "split": "train"}
{"id": "6d89a7320d_4", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "6d89a7320d_5", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. window sill\nB. keyboard\nC. monitor\nD. telephone", "choices": ["A. window sill", "B. keyboard", "C. monitor", "D. telephone"], "answer": "C. monitor", "target": "power strip", "split": "train"}
{"id": "6d89a7320d_6", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also near the keyboard, about 23 cm away. It is further near the window sill, about 30 cm away.", "target": "power strip", "split": "train"}
{"id": "6d89a7320d_7", "scene_id": "6d89a7320d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 23.1}, {\"object\": \"window sill\", \"relation\": \"near\", \"distance_cm\": 29.8}]}", "target": "power strip", "split": "train"}
{"id": "6da1d5ab04_0", "scene_id": "6da1d5ab04", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "6da1d5ab04_1", "scene_id": "6da1d5ab04", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote control in this scene?\nA. lamp\nB. backpack\nC. chair\nD. mirror", "choices": ["A. lamp", "B. backpack", "C. chair", "D. mirror"], "answer": "C. chair", "target": "remote control", "split": "train"}
{"id": "6da1d5ab04_2", "scene_id": "6da1d5ab04", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote control in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote control is on the table. It is also to the side of and above the chair, about 18 cm away. It is further to the side of and below the lamp, about 21 cm away.", "target": "remote control", "split": "train"}
{"id": "6da1d5ab04_3", "scene_id": "6da1d5ab04", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote control in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote control\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 17.9}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 20.7}]}", "target": "remote control", "split": "train"}
{"id": "6ea60e282a_0", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "6ea60e282a_1", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. cream tube\nB. book\nC. office chair\nD. paper towel", "choices": ["A. cream tube", "B. book", "C. office chair", "D. paper towel"], "answer": "D. paper towel", "target": "cup", "split": "train"}
{"id": "6ea60e282a_2", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the cabinet. It is also next to the paper towel, about 9 cm away. It is further to the side of and below the book, about 24 cm away.", "target": "cup", "split": "train"}
{"id": "6ea60e282a_3", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"paper towel\", \"relation\": \"next_to\", \"distance_cm\": 8.9}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 24.2}]}", "target": "cup", "split": "train"}
{"id": "6ea60e282a_4", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "6ea60e282a_5", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. file organizer\nB. monitor\nC. office chair\nD. sticky notes", "choices": ["A. file organizer", "B. monitor", "C. office chair", "D. sticky notes"], "answer": "B. monitor", "target": "keyboard", "split": "train"}
{"id": "6ea60e282a_6", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the monitor, about 11 cm away. It is also near the sticky notes, about 23 cm away. It is further to the side of and above the office chair, about 25 cm away.", "target": "keyboard", "split": "train"}
{"id": "6ea60e282a_7", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.6}, {\"object\": \"sticky notes\", \"relation\": \"near\", \"distance_cm\": 22.6}]}", "target": "keyboard", "split": "train"}
{"id": "6ea60e282a_8", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "No", "target": "phone charger", "split": "train"}
{"id": "6ea60e282a_9", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. file organizer\nB. sticky notes\nC. wrist rest\nD. monitor", "choices": ["A. file organizer", "B. sticky notes", "C. wrist rest", "D. monitor"], "answer": "C. wrist rest", "target": "mouse", "split": "train"}
{"id": "6ea60e282a_10", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the wrist rest, about 4 cm away. It is also to the side of and below the monitor, about 15 cm away. It is further near the sticky notes, about 19 cm away.", "target": "mouse", "split": "train"}
{"id": "6ea60e282a_11", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"wrist rest\", \"relation\": \"next_to\", \"distance_cm\": 4.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 14.9}]}", "target": "mouse", "split": "train"}
{"id": "6ea60e282a_12", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "6ea60e282a_13", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. office chair\nB. sticky notes\nC. file organizer\nD. monitor", "choices": ["A. office chair", "B. sticky notes", "C. file organizer", "D. monitor"], "answer": "C. file organizer", "target": "mousepad", "split": "train"}
{"id": "6ea60e282a_14", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is to the side of and below the file organizer, about 9 cm away. It is also to the side of and below the monitor, about 13 cm away. It is further to the side of and below the sticky notes, about 14 cm away.", "target": "mousepad", "split": "train"}
{"id": "6ea60e282a_15", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"file organizer\", \"relation\": \"side_below\", \"distance_cm\": 9.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.6}]}", "target": "mousepad", "split": "train"}
{"id": "6ea60e282a_16", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "6ea60e282a_17", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. book\nB. paper\nC. cabinet\nD. stationery", "choices": ["A. book", "B. paper", "C. cabinet", "D. stationery"], "answer": "C. cabinet", "target": "pen", "split": "train"}
{"id": "6ea60e282a_18", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is to the side of and above the cabinet, about 6 cm away. It is also to the side of and above the book, about 11 cm away. It is further to the side of and above the stationery, about 32 cm away.", "target": "pen", "split": "train"}
{"id": "6ea60e282a_19", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 5.5}, {\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 11.3}]}", "target": "pen", "split": "train"}
{"id": "6ea60e282a_20", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "6ea60e282a_21", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is on the table. It is also next to the paper. It is further to the side of and above the cabinet, about 21 cm away.", "target": "stapler", "split": "train"}
{"id": "6ea60e282a_22", "scene_id": "6ea60e282a", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 20.6}]}", "target": "stapler", "split": "train"}
{"id": "6ee2fc1070_0", "scene_id": "6ee2fc1070", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "6ee2fc1070_1", "scene_id": "6ee2fc1070", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothes\nB. bedside table\nC. chair\nD. bottle", "choices": ["A. clothes", "B. bedside table", "C. chair", "D. bottle"], "answer": "A. clothes", "target": "backpack", "split": "train"}
{"id": "6ee2fc1070_2", "scene_id": "6ee2fc1070", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the clothes, about 15 cm away.", "target": "backpack", "split": "train"}
{"id": "6ee2fc1070_3", "scene_id": "6ee2fc1070", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_above\", \"distance_cm\": 14.6}]}", "target": "backpack", "split": "train"}
{"id": "6ee2fc1070_4", "scene_id": "6ee2fc1070", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "6ee2fc1070_5", "scene_id": "6ee2fc1070", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is near the power socket, about 21 cm away. It is also to the side of and below the light switch, about 21 cm away. It is further to the side of and below the wall lamp, about 35 cm away.", "target": "bottle", "split": "train"}
{"id": "6ee2fc1070_6", "scene_id": "6ee2fc1070", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"power socket\", \"relation\": \"near\", \"distance_cm\": 21.0}, {\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 21.1}]}", "target": "bottle", "split": "train"}
{"id": "709ab5bffe_0", "scene_id": "709ab5bffe", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "709ab5bffe_1", "scene_id": "709ab5bffe", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. kitchen cabinet\nB. basket\nC. dishwasher\nD. refrigerator", "choices": ["A. kitchen cabinet", "B. basket", "C. dishwasher", "D. refrigerator"], "answer": "B. basket", "target": "bottle", "split": "train"}
{"id": "709ab5bffe_2", "scene_id": "709ab5bffe", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the basket, about 3 cm away. It is also to the side of and below the kitchen cabinet, about 20 cm away. It is further to the side of and above the dishwasher, about 31 cm away.", "target": "bottle", "split": "train"}
{"id": "709ab5bffe_3", "scene_id": "709ab5bffe", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"basket\", \"relation\": \"under\", \"distance_cm\": 2.7}, {\"object\": \"kitchen cabinet\", \"relation\": \"side_below\", \"distance_cm\": 20.4}]}", "target": "bottle", "split": "train"}
{"id": "70f0e494b2_0", "scene_id": "70f0e494b2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "70f0e494b2_1", "scene_id": "70f0e494b2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. ceiling fan\nB. monitor\nC. table\nD. shelf", "choices": ["A. ceiling fan", "B. monitor", "C. table", "D. shelf"], "answer": "D. shelf", "target": "cup", "split": "train"}
{"id": "70f0e494b2_2", "scene_id": "70f0e494b2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the shelf, about 3 cm away. It is also to the side of and below the ceiling fan, about 42 cm away.", "target": "cup", "split": "train"}
{"id": "70f0e494b2_3", "scene_id": "70f0e494b2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 3.1}, {\"object\": \"ceiling fan\", \"relation\": \"side_below\", \"distance_cm\": 41.8}]}", "target": "cup", "split": "train"}
{"id": "712dc47104_0", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "train"}
{"id": "712dc47104_1", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. chair\nB. power socket\nC. heater\nD. switch", "choices": ["A. chair", "B. power socket", "C. heater", "D. switch"], "answer": "D. switch", "target": "bottle", "split": "train"}
{"id": "712dc47104_2", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the switch, about 10 cm away. It is also to the side of and above the heater, about 35 cm away. It is further to the side of and above the chair, about 38 cm away.", "target": "bottle", "split": "train"}
{"id": "712dc47104_3", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"switch\", \"relation\": \"side_above\", \"distance_cm\": 10.3}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 34.6}]}", "target": "bottle", "split": "train"}
{"id": "712dc47104_4", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "712dc47104_5", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and above the chair, about 25 cm away. It is further to the side of and below the speaker, about 25 cm away.", "target": "mouse", "split": "train"}
{"id": "712dc47104_6", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 24.9}, {\"object\": \"speaker\", \"relation\": \"side_below\", \"distance_cm\": 24.9}]}", "target": "mouse", "split": "train"}
{"id": "712dc47104_7", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "Yes", "target": "remote", "split": "train"}
{"id": "712dc47104_8", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. mouse\nB. paper\nC. chair\nD. speaker", "choices": ["A. mouse", "B. paper", "C. chair", "D. speaker"], "answer": "B. paper", "target": "remote", "split": "train"}
{"id": "712dc47104_9", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is on the table. It is also next to the paper, about 10 cm away. It is further to the side of and below the speaker, about 20 cm away.", "target": "remote", "split": "train"}
{"id": "712dc47104_10", "scene_id": "712dc47104", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 10.2}, {\"object\": \"speaker\", \"relation\": \"side_below\", \"distance_cm\": 19.6}]}", "target": "remote", "split": "train"}
{"id": "724c40236c_0", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "Yes", "target": "mouse pad", "split": "train"}
{"id": "724c40236c_1", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. keyboard\nB. backpack\nC. monitor\nD. office chair", "choices": ["A. keyboard", "B. backpack", "C. monitor", "D. office chair"], "answer": "A. keyboard", "target": "mouse pad", "split": "train"}
{"id": "724c40236c_2", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is to the side of and above the office chair, about 13 cm away. It is also to the side of and below the monitor, about 38 cm away. It is further to the side of and above the backpack, about 40 cm away.", "target": "mouse pad", "split": "train"}
{"id": "724c40236c_3", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 38.2}]}", "target": "mouse pad", "split": "train"}
{"id": "724c40236c_4", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "724c40236c_5", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. footstool\nB. sofa\nC. backpack\nD. pillow", "choices": ["A. footstool", "B. sofa", "C. backpack", "D. pillow"], "answer": "C. backpack", "target": "notebook", "split": "train"}
{"id": "724c40236c_6", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the coffee table. It is also to the side of and above the backpack, about 17 cm away. It is further to the side of and above the footstool, about 19 cm away.", "target": "notebook", "split": "train"}
{"id": "724c40236c_7", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 17.4}, {\"object\": \"footstool\", \"relation\": \"side_above\", \"distance_cm\": 18.8}]}", "target": "notebook", "split": "train"}
{"id": "724c40236c_8", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "724c40236c_9", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the sticky note in this scene?\nA. table\nB. column\nC. toy\nD. keyboard", "choices": ["A. table", "B. column", "C. toy", "D. keyboard"], "answer": "C. toy", "target": "sticky note", "split": "train"}
{"id": "724c40236c_10", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the sticky note in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The sticky note is to the side of and above the toy, about 4 cm away. It is also to the side of and above the table, about 7 cm away. It is further to the side of and above the keyboard, about 10 cm away.", "target": "sticky note", "split": "train"}
{"id": "724c40236c_11", "scene_id": "724c40236c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the sticky note in structured JSON format.", "choices": null, "answer": "{\"target\": \"sticky note\", \"support_surface\": \"table\", \"references\": [{\"object\": \"toy\", \"relation\": \"side_above\", \"distance_cm\": 3.9}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 7.0}]}", "target": "sticky note", "split": "train"}
{"id": "725b8f0cba_0", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "725b8f0cba_1", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. books\nB. chair\nC. laptop\nD. bedsheets", "choices": ["A. books", "B. chair", "C. laptop", "D. bedsheets"], "answer": "C. laptop", "target": "bottle", "split": "train"}
{"id": "725b8f0cba_2", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the desk. It is also to the side of and above the laptop, about 5 cm away. It is further to the side of and above the books, about 8 cm away.", "target": "bottle", "split": "train"}
{"id": "725b8f0cba_3", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 5.0}, {\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 8.2}]}", "target": "bottle", "split": "train"}
{"id": "725b8f0cba_4", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "725b8f0cba_5", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. pen holder\nB. sticky note\nC. wheeled chair\nD. books", "choices": ["A. pen holder", "B. sticky note", "C. wheeled chair", "D. books"], "answer": "A. pen holder", "target": "cup", "split": "train"}
{"id": "725b8f0cba_6", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the pen holder, about 4 cm away. It is further next to the books, about 11 cm away.", "target": "cup", "split": "train"}
{"id": "725b8f0cba_7", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"pen holder\", \"relation\": \"next_to\", \"distance_cm\": 4.4}, {\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 10.9}]}", "target": "cup", "split": "train"}
{"id": "725b8f0cba_8", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "725b8f0cba_9", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the desk. It is also to the side of and above the books, about 4 cm away. It is further next to the cup, about 4 cm away.", "target": "pen holder", "split": "train"}
{"id": "725b8f0cba_10", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 3.6}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 4.4}]}", "target": "pen holder", "split": "train"}
{"id": "725b8f0cba_11", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "725b8f0cba_12", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. nightstand\nB. lamp\nC. router\nD. shelf", "choices": ["A. nightstand", "B. lamp", "C. router", "D. shelf"], "answer": "A. nightstand", "target": "power strip", "split": "train"}
{"id": "725b8f0cba_13", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the nightstand, about 3 cm away. It is also to the side of and below the lamp, about 24 cm away. It is further to the side of and below the shelf, about 37 cm away.", "target": "power strip", "split": "train"}
{"id": "725b8f0cba_14", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"nightstand\", \"relation\": \"under\", \"distance_cm\": 2.9}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 24.0}]}", "target": "power strip", "split": "train"}
{"id": "725b8f0cba_15", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "725b8f0cba_16", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. lamp\nB. nightstand\nC. mug\nD. box", "choices": ["A. lamp", "B. nightstand", "C. mug", "D. box"], "answer": "B. nightstand", "target": "tissue box", "split": "train"}
{"id": "725b8f0cba_17", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the shelf. It is also to the side of and above the nightstand, about 5 cm away. It is further to the side of and below the lamp, about 11 cm away.", "target": "tissue box", "split": "train"}
{"id": "725b8f0cba_18", "scene_id": "725b8f0cba", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"nightstand\", \"relation\": \"side_above\", \"distance_cm\": 5.1}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 11.1}]}", "target": "tissue box", "split": "train"}
{"id": "72f527a47c_0", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "72f527a47c_1", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. picture\nB. monitor\nC. chair\nD. scissors", "choices": ["A. picture", "B. monitor", "C. chair", "D. scissors"], "answer": "B. monitor", "target": "mug", "split": "train"}
{"id": "72f527a47c_2", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and below the monitor, about 9 cm away. It is also to the side of and above the chair, about 38 cm away. It is further near the scissors, about 57 cm away.", "target": "mug", "split": "train"}
{"id": "72f527a47c_3", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.5}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 38.0}]}", "target": "mug", "split": "train"}
{"id": "72f527a47c_4", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "No", "target": "highlighter", "split": "train"}
{"id": "72f527a47c_5", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scissors in this scene?\nA. monitor\nB. chair\nC. mouse\nD. keyboard", "choices": ["A. monitor", "B. chair", "C. mouse", "D. keyboard"], "answer": "A. monitor", "target": "scissors", "split": "train"}
{"id": "72f527a47c_6", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is to the side of and below the monitor, about 6 cm away. It is also near the mouse, about 16 cm away. It is further to the side of and above the chair, about 24 cm away.", "target": "scissors", "split": "train"}
{"id": "72f527a47c_7", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.6}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 15.8}]}", "target": "scissors", "split": "train"}
{"id": "72f527a47c_8", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "train"}
{"id": "72f527a47c_9", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. box\nB. kettle\nC. cabinet\nD. binders", "choices": ["A. box", "B. kettle", "C. cabinet", "D. binders"], "answer": "B. kettle", "target": "tote bag", "split": "train"}
{"id": "72f527a47c_10", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is on the table. It is also to the side of and below the kettle, about 28 cm away.", "target": "tote bag", "split": "train"}
{"id": "72f527a47c_11", "scene_id": "72f527a47c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"table\", \"references\": [{\"object\": \"kettle\", \"relation\": \"side_below\", \"distance_cm\": 27.8}]}", "target": "tote bag", "split": "train"}
{"id": "734bd1a97e_0", "scene_id": "734bd1a97e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "734bd1a97e_1", "scene_id": "734bd1a97e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. power outlet\nB. stool\nC. computer tower\nD. charger head", "choices": ["A. power outlet", "B. stool", "C. computer tower", "D. charger head"], "answer": "C. computer tower", "target": "extension cord", "split": "train"}
{"id": "734bd1a97e_2", "scene_id": "734bd1a97e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the computer tower, about 6 cm away. It is also to the side of and below the power outlet, about 9 cm away. It is further to the side of and below the charger head, about 17 cm away.", "target": "extension cord", "split": "train"}
{"id": "734bd1a97e_3", "scene_id": "734bd1a97e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_below\", \"distance_cm\": 5.8}, {\"object\": \"power outlet\", \"relation\": \"side_below\", \"distance_cm\": 9.5}]}", "target": "extension cord", "split": "train"}
{"id": "7357c3702a_0", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "7357c3702a_1", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. bedside table\nB. chair\nC. power sockets\nD. pillow", "choices": ["A. bedside table", "B. chair", "C. power sockets", "D. pillow"], "answer": "A. bedside table", "target": "extension cord", "split": "train"}
{"id": "7357c3702a_2", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the bedside table, about 7 cm away. It is also to the side of and below the pillow, about 73 cm away.", "target": "extension cord", "split": "train"}
{"id": "7357c3702a_3", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"bedside table\", \"relation\": \"side_below\", \"distance_cm\": 7.3}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 72.7}]}", "target": "extension cord", "split": "train"}
{"id": "7357c3702a_4", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "Yes", "target": "tissue", "split": "train"}
{"id": "7357c3702a_5", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue in this scene?\nA. standing mirror\nB. table\nC. chair\nD. wallet", "choices": ["A. standing mirror", "B. table", "C. chair", "D. wallet"], "answer": "B. table", "target": "tissue", "split": "train"}
{"id": "7357c3702a_6", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue is to the side of and above the table, about 6 cm away. It is also to the side of and above the wallet, about 18 cm away. It is further to the side of and above the standing mirror, about 43 cm away.", "target": "tissue", "split": "train"}
{"id": "7357c3702a_7", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 6.1}, {\"object\": \"wallet\", \"relation\": \"side_above\", \"distance_cm\": 17.6}]}", "target": "tissue", "split": "train"}
{"id": "7357c3702a_8", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "7357c3702a_9", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. power sockets\nB. wallet\nC. standing mirror\nD. chair", "choices": ["A. power sockets", "B. wallet", "C. standing mirror", "D. chair"], "answer": "B. wallet", "target": "tissue box", "split": "train"}
{"id": "7357c3702a_10", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also next to the wallet, about 13 cm away. It is further to the side of and above the standing mirror, about 38 cm away.", "target": "tissue box", "split": "train"}
{"id": "7357c3702a_11", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"wallet\", \"relation\": \"next_to\", \"distance_cm\": 13.1}, {\"object\": \"standing mirror\", \"relation\": \"side_above\", \"distance_cm\": 38.0}]}", "target": "tissue box", "split": "train"}
{"id": "7357c3702a_12", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "7357c3702a_13", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the wallet in this scene?\nA. tissue box\nB. tissue\nC. chair\nD. standing mirror", "choices": ["A. tissue box", "B. tissue", "C. chair", "D. standing mirror"], "answer": "A. tissue box", "target": "wallet", "split": "train"}
{"id": "7357c3702a_14", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the wallet in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The wallet is on the table. It is also next to the tissue box, about 13 cm away. It is further to the side of and below the tissue, about 18 cm away.", "target": "wallet", "split": "train"}
{"id": "7357c3702a_15", "scene_id": "7357c3702a", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the wallet in structured JSON format.", "choices": null, "answer": "{\"target\": \"wallet\", \"support_surface\": \"table\", \"references\": [{\"object\": \"tissue box\", \"relation\": \"next_to\", \"distance_cm\": 13.1}, {\"object\": \"tissue\", \"relation\": \"side_below\", \"distance_cm\": 17.6}]}", "target": "wallet", "split": "train"}
{"id": "75d29d69b8_0", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "75d29d69b8_1", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cup\nB. keyboard\nC. heater\nD. monitor", "choices": ["A. cup", "B. keyboard", "C. heater", "D. monitor"], "answer": "C. heater", "target": "backpack", "split": "train"}
{"id": "75d29d69b8_2", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the window, about 7 cm away. It is also to the side of and above the heater, about 23 cm away.", "target": "backpack", "split": "train"}
{"id": "75d29d69b8_3", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"window\", \"references\": [{\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 6.7}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 23.4}]}", "target": "backpack", "split": "train"}
{"id": "75d29d69b8_4", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "75d29d69b8_5", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. monitor\nB. backpack\nC. keyboard\nD. heater", "choices": ["A. monitor", "B. backpack", "C. keyboard", "D. heater"], "answer": "A. monitor", "target": "cup", "split": "train"}
{"id": "75d29d69b8_6", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also near the keyboard, about 16 cm away.", "target": "cup", "split": "train"}
{"id": "75d29d69b8_7", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 16.4}]}", "target": "cup", "split": "train"}
{"id": "75d29d69b8_8", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "75d29d69b8_9", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. heater\nC. cup\nD. backpack", "choices": ["A. monitor", "B. heater", "C. cup", "D. backpack"], "answer": "A. monitor", "target": "keyboard", "split": "train"}
{"id": "75d29d69b8_10", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is under the monitor, about 5 cm away. It is also near the cup, about 16 cm away.", "target": "keyboard", "split": "train"}
{"id": "75d29d69b8_11", "scene_id": "75d29d69b8", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.5}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 16.4}]}", "target": "keyboard", "split": "train"}
{"id": "7692b88644_0", "scene_id": "7692b88644", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "7692b88644_1", "scene_id": "7692b88644", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. sword\nB. weighted plate\nC. weighted plates\nD. mop", "choices": ["A. sword", "B. weighted plate", "C. weighted plates", "D. mop"], "answer": "D. mop", "target": "backpack", "split": "train"}
{"id": "7692b88644_2", "scene_id": "7692b88644", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is under the mop. It is also to the side of and below the sword, about 23 cm away. It is further to the side of and below the weighted plates, about 64 cm away.", "target": "backpack", "split": "train"}
{"id": "7692b88644_3", "scene_id": "7692b88644", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"rope\", \"references\": [{\"object\": \"mop\", \"relation\": \"under\", \"distance_cm\": 2.3}, {\"object\": \"sword\", \"relation\": \"side_below\", \"distance_cm\": 22.7}]}", "target": "backpack", "split": "train"}
{"id": "76f0eb2577_0", "scene_id": "76f0eb2577", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "76f0eb2577_1", "scene_id": "76f0eb2577", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. laptop\nB. office chair\nC. monitor\nD. laptop charger", "choices": ["A. laptop", "B. office chair", "C. monitor", "D. laptop charger"], "answer": "D. laptop charger", "target": "extension cord", "split": "train"}
{"id": "76f0eb2577_2", "scene_id": "76f0eb2577", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the desk. It is also to the side of and below the monitor, about 14 cm away. It is further near the laptop, about 21 cm away.", "target": "extension cord", "split": "train"}
{"id": "76f0eb2577_3", "scene_id": "76f0eb2577", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.9}, {\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 20.7}]}", "target": "extension cord", "split": "train"}
{"id": "7739004a45_0", "scene_id": "7739004a45", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "7739004a45_1", "scene_id": "7739004a45", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. toilet\nB. toilet brush\nC. toilet paper\nD. mop", "choices": ["A. toilet", "B. toilet brush", "C. toilet paper", "D. mop"], "answer": "B. toilet brush", "target": "bottle", "split": "train"}
{"id": "7739004a45_2", "scene_id": "7739004a45", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the toilet brush, about 3 cm away. It is also to the side of and below the mop, about 19 cm away. It is further to the side of and below the toilet, about 24 cm away.", "target": "bottle", "split": "train"}
{"id": "7739004a45_3", "scene_id": "7739004a45", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet brush\", \"relation\": \"under\", \"distance_cm\": 2.5}, {\"object\": \"mop\", \"relation\": \"side_below\", \"distance_cm\": 19.2}]}", "target": "bottle", "split": "train"}
{"id": "77fe971eba_0", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "77fe971eba_1", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. wine bottle\nB. pillow\nC. yoga mat\nD. thermos", "choices": ["A. wine bottle", "B. pillow", "C. yoga mat", "D. thermos"], "answer": "A. wine bottle", "target": "backpack", "split": "train"}
{"id": "77fe971eba_2", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the wine bottle, about 9 cm away. It is also to the side of and below the yoga mat, about 11 cm away. It is further to the side of and below the thermos, about 26 cm away.", "target": "backpack", "split": "train"}
{"id": "77fe971eba_3", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"wine bottle\", \"relation\": \"side_below\", \"distance_cm\": 9.4}, {\"object\": \"yoga mat\", \"relation\": \"side_below\", \"distance_cm\": 11.5}]}", "target": "backpack", "split": "train"}
{"id": "77fe971eba_4", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "77fe971eba_5", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. notebooks\nB. pen holder\nC. curtain\nD. table", "choices": ["A. notebooks", "B. pen holder", "C. curtain", "D. table"], "answer": "D. table", "target": "mouse", "split": "train"}
{"id": "77fe971eba_6", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the table, about 5 cm away. It is also to the side of and below the curtain, about 7 cm away. It is further next to the notebooks, about 8 cm away.", "target": "mouse", "split": "train"}
{"id": "77fe971eba_7", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.8}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 6.8}]}", "target": "mouse", "split": "train"}
{"id": "77fe971eba_8", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "77fe971eba_9", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. power socket\nB. laptop bag\nC. tv remote\nD. light switch", "choices": ["A. power socket", "B. laptop bag", "C. tv remote", "D. light switch"], "answer": "C. tv remote", "target": "mousepad", "split": "train"}
{"id": "77fe971eba_10", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also next to the tv remote. It is further to the side of and above the laptop bag, about 19 cm away.", "target": "mousepad", "split": "train"}
{"id": "77fe971eba_11", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"tv remote\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"laptop bag\", \"relation\": \"side_above\", \"distance_cm\": 19.4}]}", "target": "mousepad", "split": "train"}
{"id": "77fe971eba_12", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "77fe971eba_13", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. laptop\nB. laptop bag\nC. tv remote\nD. mouse", "choices": ["A. laptop", "B. laptop bag", "C. tv remote", "D. mouse"], "answer": "D. mouse", "target": "pen holder", "split": "train"}
{"id": "77fe971eba_14", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also to the side of and above the mouse, about 8 cm away. It is further next to the laptop, about 11 cm away.", "target": "pen holder", "split": "train"}
{"id": "77fe971eba_15", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"side_above\", \"distance_cm\": 8.1}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 11.1}]}", "target": "pen holder", "split": "train"}
{"id": "77fe971eba_16", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "77fe971eba_17", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. pen holder\nB. mousepad\nC. notebooks\nD. mouse", "choices": ["A. pen holder", "B. mousepad", "C. notebooks", "D. mouse"], "answer": "B. mousepad", "target": "tv remote", "split": "train"}
{"id": "77fe971eba_18", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the table. It is also next to the mousepad. It is further next to the notebooks, about 12 cm away.", "target": "tv remote", "split": "train"}
{"id": "77fe971eba_19", "scene_id": "77fe971eba", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"notebooks\", \"relation\": \"next_to\", \"distance_cm\": 12.2}]}", "target": "tv remote", "split": "train"}
{"id": "781f5de0a8_0", "scene_id": "781f5de0a8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "781f5de0a8_1", "scene_id": "781f5de0a8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is next to the box, about 9 cm away. It is also to the side of and below the air purifier, about 10 cm away. It is further to the side of and below the bag, about 19 cm away.", "target": "extension cord", "split": "train"}
{"id": "781f5de0a8_2", "scene_id": "781f5de0a8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 9.2}, {\"object\": \"air purifier\", \"relation\": \"side_below\", \"distance_cm\": 9.6}]}", "target": "extension cord", "split": "train"}
{"id": "781f5de0a8_3", "scene_id": "781f5de0a8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "781f5de0a8_4", "scene_id": "781f5de0a8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. chair\nB. slippers\nC. air purifier\nD. bag", "choices": ["A. chair", "B. slippers", "C. air purifier", "D. bag"], "answer": "A. chair", "target": "spray bottle", "split": "train"}
{"id": "781f5de0a8_5", "scene_id": "781f5de0a8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and below the chair, about 6 cm away. It is also to the side of and above the slippers, about 18 cm away.", "target": "spray bottle", "split": "train"}
{"id": "781f5de0a8_6", "scene_id": "781f5de0a8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 5.9}, {\"object\": \"slippers\", \"relation\": \"side_above\", \"distance_cm\": 17.7}]}", "target": "spray bottle", "split": "train"}
{"id": "79eb3fbc63_0", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tape dispenser appear?", "choices": null, "answer": "No", "target": "tape dispenser", "split": "train"}
{"id": "79eb3fbc63_1", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. desk\nB. mug\nC. cup\nD. coax outlets", "choices": ["A. desk", "B. mug", "C. cup", "D. coax outlets"], "answer": "D. coax outlets", "target": "charger", "split": "train"}
{"id": "79eb3fbc63_2", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is next to the coax outlets, about 8 cm away. It is also to the side of and below the desk, about 31 cm away. It is further to the side of and below the cup, about 53 cm away.", "target": "charger", "split": "train"}
{"id": "79eb3fbc63_3", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"coax outlets\", \"relation\": \"next_to\", \"distance_cm\": 7.8}, {\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 31.5}]}", "target": "charger", "split": "train"}
{"id": "79eb3fbc63_4", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "79eb3fbc63_5", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. coax outlets\nB. charger\nC. power sockets\nD. tea box", "choices": ["A. coax outlets", "B. charger", "C. power sockets", "D. tea box"], "answer": "D. tea box", "target": "cup", "split": "train"}
{"id": "79eb3fbc63_6", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the tea box, about 6 cm away. It is further to the side of and above the coax outlets, about 50 cm away.", "target": "cup", "split": "train"}
{"id": "79eb3fbc63_7", "scene_id": "79eb3fbc63", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"tea box\", \"relation\": \"next_to\", \"distance_cm\": 6.0}, {\"object\": \"coax outlets\", \"relation\": \"side_above\", \"distance_cm\": 50.0}]}", "target": "cup", "split": "train"}
{"id": "7a0a669e90_0", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "7a0a669e90_1", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. keyboard\nB. chair\nC. book\nD. office table", "choices": ["A. keyboard", "B. chair", "C. book", "D. office table"], "answer": "D. office table", "target": "headphones", "split": "train"}
{"id": "7a0a669e90_2", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the office table. It is also next to the keyboard, about 14 cm away. It is further to the side of and above the chair, about 15 cm away.", "target": "headphones", "split": "train"}
{"id": "7a0a669e90_3", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 2.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 14.0}]}", "target": "headphones", "split": "train"}
{"id": "7a0a669e90_4", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "7a0a669e90_5", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. book\nB. headphones\nC. mouse\nD. monitor", "choices": ["A. book", "B. headphones", "C. mouse", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "7a0a669e90_6", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is under the monitor, about 4 cm away. It is also next to the book, about 13 cm away. It is further next to the headphones, about 14 cm away.", "target": "keyboard", "split": "train"}
{"id": "7a0a669e90_7", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.3}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 13.1}]}", "target": "keyboard", "split": "train"}
{"id": "7a0a669e90_8", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "7a0a669e90_9", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. telephone\nD. cabinet", "choices": ["A. keyboard", "B. monitor", "C. telephone", "D. cabinet"], "answer": "C. telephone", "target": "mouse", "split": "train"}
{"id": "7a0a669e90_10", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the telephone, about 7 cm away. It is also next to the keyboard, about 15 cm away. It is further to the side of and below the monitor, about 20 cm away.", "target": "mouse", "split": "train"}
{"id": "7a0a669e90_11", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 6.6}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 15.0}]}", "target": "mouse", "split": "train"}
{"id": "7a0a669e90_12", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "7a0a669e90_13", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. dustbin\nB. cabinet\nC. book\nD. books", "choices": ["A. dustbin", "B. cabinet", "C. book", "D. books"], "answer": "A. dustbin", "target": "power adapter", "split": "train"}
{"id": "7a0a669e90_14", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and below the door, about 11 cm away. It is also near the dustbin, about 66 cm away.", "target": "power adapter", "split": "train"}
{"id": "7a0a669e90_15", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 10.8}, {\"object\": \"dustbin\", \"relation\": \"near\", \"distance_cm\": 66.5}]}", "target": "power adapter", "split": "train"}
{"id": "7a0a669e90_16", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "7a0a669e90_17", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. heater\nB. book\nC. office table\nD. keyboard", "choices": ["A. heater", "B. book", "C. office table", "D. keyboard"], "answer": "C. office table", "target": "power strip", "split": "train"}
{"id": "7a0a669e90_18", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the office table, about 15 cm away. It is also to the side of and below the heater, about 40 cm away. It is further to the side of and below the book, about 75 cm away.", "target": "power strip", "split": "train"}
{"id": "7a0a669e90_19", "scene_id": "7a0a669e90", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_below\", \"distance_cm\": 14.7}, {\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 39.8}]}", "target": "power strip", "split": "train"}
{"id": "7a7980bb69_0", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "7a7980bb69_1", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "7a7980bb69_2", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. air purifier\nB. carrying case\nC. tall table\nD. clothes drying rack", "choices": ["A. air purifier", "B. carrying case", "C. tall table", "D. clothes drying rack"], "answer": "C. tall table", "target": "cup", "split": "train"}
{"id": "7a7980bb69_3", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is under the tall table, about 4 cm away. It is also to the side of and below the door, about 9 cm away. It is further to the side of and below the clothes drying rack, about 14 cm away.", "target": "cup", "split": "train"}
{"id": "7a7980bb69_4", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"tall table\", \"references\": [{\"object\": \"tall table\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 9.3}]}", "target": "cup", "split": "train"}
{"id": "7a7980bb69_5", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "No", "target": "bottle", "split": "train"}
{"id": "7a7980bb69_6", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the tall table, about 8 cm away. It is also to the side of and below the radiator, about 8 cm away. It is further to the side of and below the kitchen cabinet, about 16 cm away.", "target": "extension cord", "split": "train"}
{"id": "7a7980bb69_7", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"tall table\", \"references\": [{\"object\": \"tall table\", \"relation\": \"side_below\", \"distance_cm\": 8.3}, {\"object\": \"radiator\", \"relation\": \"side_below\", \"distance_cm\": 8.3}]}", "target": "extension cord", "split": "train"}
{"id": "7a7980bb69_8", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "7a7980bb69_9", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. paper towel pack\nB. plastic container\nC. radiator\nD. extention cord", "choices": ["A. paper towel pack", "B. plastic container", "C. radiator", "D. extention cord"], "answer": "B. plastic container", "target": "headphones", "split": "train"}
{"id": "7a7980bb69_10", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the plastic container, about 11 cm away. It is also near the extention cord, about 17 cm away. It is further near the paper towel pack, about 20 cm away.", "target": "headphones", "split": "train"}
{"id": "7a7980bb69_11", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"plastic container\", \"relation\": \"next_to\", \"distance_cm\": 10.9}, {\"object\": \"extention cord\", \"relation\": \"near\", \"distance_cm\": 16.6}]}", "target": "headphones", "split": "train"}
{"id": "7a7980bb69_12", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "7a7980bb69_13", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. office chair\nB. desk\nC. radiator\nD. electronic device", "choices": ["A. office chair", "B. desk", "C. radiator", "D. electronic device"], "answer": "B. desk", "target": "laptop charger", "split": "train"}
{"id": "7a7980bb69_14", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the desk, about 5 cm away. It is also to the side of and below the radiator, about 23 cm away. It is further to the side of and below the office chair, about 26 cm away.", "target": "laptop charger", "split": "train"}
{"id": "7a7980bb69_15", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 5.3}, {\"object\": \"radiator\", \"relation\": \"side_below\", \"distance_cm\": 23.3}]}", "target": "laptop charger", "split": "train"}
{"id": "7a7980bb69_16", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "7a7980bb69_17", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. radiator\nB. office chair\nC. cable raceway\nD. laptop", "choices": ["A. radiator", "B. office chair", "C. cable raceway", "D. laptop"], "answer": "D. laptop", "target": "mouse", "split": "train"}
{"id": "7a7980bb69_18", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the laptop, about 11 cm away. It is further to the side of and above the radiator, about 21 cm away.", "target": "mouse", "split": "train"}
{"id": "7a7980bb69_19", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 10.6}, {\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 20.7}]}", "target": "mouse", "split": "train"}
{"id": "7a7980bb69_20", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "7a7980bb69_21", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. paper towel\nB. trash bag pack\nC. cable raceway\nD. wall mounted organizer", "choices": ["A. paper towel", "B. trash bag pack", "C. cable raceway", "D. wall mounted organizer"], "answer": "D. wall mounted organizer", "target": "phone charger", "split": "train"}
{"id": "7a7980bb69_22", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is under the wall mounted organizer. It is also under the cable raceway, about 5 cm away. It is further to the side of and below the paper towel, about 12 cm away.", "target": "phone charger", "split": "train"}
{"id": "7a7980bb69_23", "scene_id": "7a7980bb69", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"wall mounted organizer\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"cable raceway\", \"relation\": \"under\", \"distance_cm\": 4.9}]}", "target": "phone charger", "split": "train"}
{"id": "7b4cb756d4_0", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "7b4cb756d4_1", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. tv stand\nB. tv remote\nC. sofa\nD. table", "choices": ["A. tv stand", "B. tv remote", "C. sofa", "D. table"], "answer": "A. tv stand", "target": "power strip", "split": "train"}
{"id": "7b4cb756d4_2", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the tv stand, about 7 cm away.", "target": "power strip", "split": "train"}
{"id": "7b4cb756d4_3", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"tv stand\", \"references\": [{\"object\": \"tv stand\", \"relation\": \"side_below\", \"distance_cm\": 6.8}]}", "target": "power strip", "split": "train"}
{"id": "7b4cb756d4_4", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "Yes", "target": "tv remote", "split": "train"}
{"id": "7b4cb756d4_5", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. power strip\nB. sofa\nC. tv stand\nD. table", "choices": ["A. power strip", "B. sofa", "C. tv stand", "D. table"], "answer": "B. sofa", "target": "tv remote", "split": "train"}
{"id": "7b4cb756d4_6", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is near the sofa, about 48 cm away.", "target": "tv remote", "split": "train"}
{"id": "7b4cb756d4_7", "scene_id": "7b4cb756d4", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"table\", \"references\": [{\"object\": \"sofa\", \"relation\": \"near\", \"distance_cm\": 47.9}]}", "target": "tv remote", "split": "train"}
{"id": "7b6477cb95_0", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "7b6477cb95_1", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. cup\nC. pen\nD. keyboard", "choices": ["A. monitor", "B. cup", "C. pen", "D. keyboard"], "answer": "A. monitor", "target": "bottle", "split": "train"}
{"id": "7b6477cb95_2", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and below the monitor, about 5 cm away. It is further next to the keyboard, about 8 cm away.", "target": "bottle", "split": "train"}
{"id": "7b6477cb95_3", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.4}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.4}]}", "target": "bottle", "split": "train"}
{"id": "7b6477cb95_4", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "7b6477cb95_5", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. bottle\nB. pens\nC. monitor\nD. keyboard", "choices": ["A. bottle", "B. pens", "C. monitor", "D. keyboard"], "answer": "A. bottle", "target": "cup", "split": "train"}
{"id": "7b6477cb95_6", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also next to the bottle, about 10 cm away. It is further to the side of and below the monitor, about 14 cm away.", "target": "cup", "split": "train"}
{"id": "7b6477cb95_7", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 10.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.8}]}", "target": "cup", "split": "train"}
{"id": "7b6477cb95_8", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "No", "target": "soap", "split": "train"}
{"id": "7b6477cb95_9", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. telephone\nB. book\nC. cabinet\nD. monitor", "choices": ["A. telephone", "B. book", "C. cabinet", "D. monitor"], "answer": "D. monitor", "target": "mouse", "split": "train"}
{"id": "7b6477cb95_10", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and below the monitor, about 5 cm away. It is further near the book, about 17 cm away.", "target": "mouse", "split": "train"}
{"id": "7b6477cb95_11", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.4}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 17.4}]}", "target": "mouse", "split": "train"}
{"id": "7b6477cb95_12", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "7b6477cb95_13", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor\nB. cable\nC. keyboard\nD. table", "choices": ["A. monitor", "B. cable", "C. keyboard", "D. table"], "answer": "D. table", "target": "power strip", "split": "train"}
{"id": "7b6477cb95_14", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also next to the cable, about 10 cm away. It is further to the side of and below the monitor, about 14 cm away.", "target": "power strip", "split": "train"}
{"id": "7b6477cb95_15", "scene_id": "7b6477cb95", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cable\", \"relation\": \"next_to\", \"distance_cm\": 9.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.7}]}", "target": "power strip", "split": "train"}
{"id": "7becc8122c_0", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "7becc8122c_1", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. notebook\nC. towel\nD. laptop", "choices": ["A. chair", "B. notebook", "C. towel", "D. laptop"], "answer": "B. notebook", "target": "mouse", "split": "train"}
{"id": "7becc8122c_2", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the notebook, about 7 cm away. It is further to the side of and above the towel, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "7becc8122c_3", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 6.9}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 13.9}]}", "target": "mouse", "split": "train"}
{"id": "7becc8122c_4", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "7becc8122c_5", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the desk. It is also to the side of and above the chair, about 7 cm away. It is further to the side of and above the towel, about 7 cm away.", "target": "notebook", "split": "train"}
{"id": "7becc8122c_6", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 6.6}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 6.6}]}", "target": "notebook", "split": "train"}
{"id": "7becc8122c_7", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "7becc8122c_8", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. acoustic guitar case\nB. acoustic guitar\nC. pillow\nD. shelf rail", "choices": ["A. acoustic guitar case", "B. acoustic guitar", "C. pillow", "D. shelf rail"], "answer": "D. shelf rail", "target": "phone charger", "split": "train"}
{"id": "7becc8122c_9", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the shelf rail, about 30 cm away. It is also to the side of and above the pillow, about 56 cm away.", "target": "phone charger", "split": "train"}
{"id": "7becc8122c_10", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"shelf rail\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 29.6}, {\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 56.0}]}", "target": "phone charger", "split": "train"}
{"id": "7becc8122c_11", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "7becc8122c_12", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone stand in this scene?\nA. curtain\nB. notebook\nC. rice cooker\nD. heater", "choices": ["A. curtain", "B. notebook", "C. rice cooker", "D. heater"], "answer": "B. notebook", "target": "phone stand", "split": "train"}
{"id": "7becc8122c_13", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone stand in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone stand is on the desk. It is also next to the notebook, about 11 cm away. It is further to the side of and below the rice cooker, about 22 cm away.", "target": "phone stand", "split": "train"}
{"id": "7becc8122c_14", "scene_id": "7becc8122c", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone stand in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone stand\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 11.4}, {\"object\": \"rice cooker\", \"relation\": \"side_below\", \"distance_cm\": 22.0}]}", "target": "phone stand", "split": "train"}
{"id": "7c0ba828a9_0", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "7c0ba828a9_1", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. office chair\nB. board\nC. basket\nD. box", "choices": ["A. office chair", "B. board", "C. basket", "D. box"], "answer": "A. office chair", "target": "backpack", "split": "train"}
{"id": "7c0ba828a9_2", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the office chair, about 16 cm away. It is also to the side of and above the box, about 64 cm away.", "target": "backpack", "split": "train"}
{"id": "7c0ba828a9_3", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 16.5}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 63.6}]}", "target": "backpack", "split": "train"}
{"id": "7c0ba828a9_4", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "7c0ba828a9_5", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is below the shelf, about 6 cm away. It is also below the basket, about 7 cm away. It is further above the foodstuffs, about 25 cm away.", "target": "bottle", "split": "train"}
{"id": "7c0ba828a9_6", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"below\", \"distance_cm\": 6.3}, {\"object\": \"basket\", \"relation\": \"below\", \"distance_cm\": 6.8}]}", "target": "bottle", "split": "train"}
{"id": "7c0ba828a9_7", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "7c0ba828a9_8", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. toothpaste\nB. cabinet\nC. books\nD. electric kettle", "choices": ["A. toothpaste", "B. cabinet", "C. books", "D. electric kettle"], "answer": "D. electric kettle", "target": "extension cord", "split": "train"}
{"id": "7c0ba828a9_9", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the desk. It is also to the side of and below the electric kettle, about 9 cm away. It is further to the side of and below the toothpaste, about 11 cm away.", "target": "extension cord", "split": "train"}
{"id": "7c0ba828a9_10", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"electric kettle\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"toothpaste\", \"relation\": \"side_below\", \"distance_cm\": 11.4}]}", "target": "extension cord", "split": "train"}
{"id": "7c0ba828a9_11", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "7c0ba828a9_12", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. electric kettle\nB. electric toothbrush\nC. toothpaste\nD. cabinet", "choices": ["A. electric kettle", "B. electric toothbrush", "C. toothpaste", "D. cabinet"], "answer": "B. electric toothbrush", "target": "phone charger", "split": "train"}
{"id": "7c0ba828a9_13", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is on the desk. It is also next to the electric toothbrush, about 6 cm away. It is further next to the electric kettle, about 9 cm away.", "target": "phone charger", "split": "train"}
{"id": "7c0ba828a9_14", "scene_id": "7c0ba828a9", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"electric toothbrush\", \"relation\": \"next_to\", \"distance_cm\": 6.4}, {\"object\": \"electric kettle\", \"relation\": \"next_to\", \"distance_cm\": 9.4}]}", "target": "phone charger", "split": "train"}
{"id": "7c31bccde5_0", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "7c31bccde5_1", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. shampoo bottle\nB. cooking pot\nC. cooking pot lid\nD. documents", "choices": ["A. shampoo bottle", "B. cooking pot", "C. cooking pot lid", "D. documents"], "answer": "D. documents", "target": "backpack", "split": "train"}
{"id": "7c31bccde5_2", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the documents, about 27 cm away. It is also to the side of and below the cooking pot lid, about 40 cm away. It is further to the side of and below the cooking pot, about 51 cm away.", "target": "backpack", "split": "train"}
{"id": "7c31bccde5_3", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"side table\", \"references\": [{\"object\": \"documents\", \"relation\": \"side_below\", \"distance_cm\": 26.5}, {\"object\": \"cooking pot lid\", \"relation\": \"side_below\", \"distance_cm\": 40.3}]}", "target": "backpack", "split": "train"}
{"id": "7c31bccde5_4", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "7c31bccde5_5", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. router\nB. monitor\nC. power outlet\nD. desk lamp", "choices": ["A. router", "B. monitor", "C. power outlet", "D. desk lamp"], "answer": "B. monitor", "target": "extension cord", "split": "train"}
{"id": "7c31bccde5_6", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the monitor, about 22 cm away. It is also to the side of and below the window, about 27 cm away. It is further to the side of and above the power outlet, about 45 cm away.", "target": "extension cord", "split": "train"}
{"id": "7c31bccde5_7", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk lamp\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 21.6}, {\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 27.4}]}", "target": "extension cord", "split": "train"}
{"id": "7c31bccde5_8", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "7c31bccde5_9", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. desk lamp\nB. mug\nC. table\nD. monitor", "choices": ["A. desk lamp", "B. mug", "C. table", "D. monitor"], "answer": "C. table", "target": "mouse", "split": "train"}
{"id": "7c31bccde5_10", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is above the table, about 3 cm away. It is also to the side of and below the desk lamp, about 28 cm away. It is further near the mug, about 32 cm away.", "target": "mouse", "split": "train"}
{"id": "7c31bccde5_11", "scene_id": "7c31bccde5", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"above\", \"distance_cm\": 3.3}, {\"object\": \"desk lamp\", \"relation\": \"side_below\", \"distance_cm\": 28.0}]}", "target": "mouse", "split": "train"}
{"id": "7cd2ac43b4_0", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "7cd2ac43b4_1", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. telephone\nB. cabinet\nC. bottle\nD. chair", "choices": ["A. telephone", "B. cabinet", "C. bottle", "D. chair"], "answer": "B. cabinet", "target": "backpack", "split": "train"}
{"id": "7cd2ac43b4_2", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the telephone, about 38 cm away. It is also to the side of and below the chair, about 40 cm away. It is further to the side of and below the bottle, about 52 cm away.", "target": "backpack", "split": "train"}
{"id": "7cd2ac43b4_3", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"telephone\", \"relation\": \"side_below\", \"distance_cm\": 38.0}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 39.7}]}", "target": "backpack", "split": "train"}
{"id": "7cd2ac43b4_4", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "7cd2ac43b4_5", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. cabinet\nB. backpack\nC. telephone\nD. metal board", "choices": ["A. cabinet", "B. backpack", "C. telephone", "D. metal board"], "answer": "A. cabinet", "target": "bottle", "split": "train"}
{"id": "7cd2ac43b4_6", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the cabinet, about 12 cm away. It is also to the side of and below the metal board, about 37 cm away. It is further to the side of and above the telephone, about 38 cm away.", "target": "bottle", "split": "train"}
{"id": "7cd2ac43b4_7", "scene_id": "7cd2ac43b4", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 12.0}, {\"object\": \"metal board\", \"relation\": \"side_below\", \"distance_cm\": 36.8}]}", "target": "bottle", "split": "train"}
{"id": "7d87b85065_0", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "7d87b85065_1", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "7d87b85065_2", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. mouse\nB. cabinet\nC. monitor\nD. documents", "choices": ["A. mouse", "B. cabinet", "C. monitor", "D. documents"], "answer": "D. documents", "target": "headphones", "split": "train"}
{"id": "7d87b85065_3", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and below the monitor, about 23 cm away. It is also near the mouse, about 39 cm away. It is further to the side of and above the cabinet, about 52 cm away.", "target": "headphones", "split": "train"}
{"id": "7d87b85065_4", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 22.6}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 38.8}]}", "target": "headphones", "split": "train"}
{"id": "7d87b85065_5", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "7d87b85065_6", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. mouse\nC. documents\nD. monitor", "choices": ["A. chair", "B. mouse", "C. documents", "D. monitor"], "answer": "C. documents", "target": "keyboard", "split": "train"}
{"id": "7d87b85065_7", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the documents, about 7 cm away. It is further next to the mouse, about 9 cm away.", "target": "keyboard", "split": "train"}
{"id": "7d87b85065_8", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"documents\", \"relation\": \"next_to\", \"distance_cm\": 7.2}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 8.9}]}", "target": "keyboard", "split": "train"}
{"id": "7d87b85065_9", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "7d87b85065_10", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. keyboard\nC. telephone\nD. file", "choices": ["A. monitor", "B. keyboard", "C. telephone", "D. file"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "7d87b85065_11", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 9 cm away. It is further to the side of and below the monitor, about 15 cm away.", "target": "mouse", "split": "train"}
{"id": "7d87b85065_12", "scene_id": "7d87b85065", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 14.8}]}", "target": "mouse", "split": "train"}
{"id": "7d8d37ca38_0", "scene_id": "7d8d37ca38", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "7d8d37ca38_1", "scene_id": "7d8d37ca38", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. heater\nB. controller\nC. pipe\nD. windowsill", "choices": ["A. heater", "B. controller", "C. pipe", "D. windowsill"], "answer": "D. windowsill", "target": "cable", "split": "train"}
{"id": "7d8d37ca38_2", "scene_id": "7d8d37ca38", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and below the windowsill, about 7 cm away. It is also to the side of and below the door, about 24 cm away. It is further to the side of and below the controller, about 33 cm away.", "target": "cable", "split": "train"}
{"id": "7d8d37ca38_3", "scene_id": "7d8d37ca38", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"door\", \"references\": [{\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 6.9}, {\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 24.1}]}", "target": "cable", "split": "train"}
{"id": "7dab70c8c8_0", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "train"}
{"id": "7dab70c8c8_1", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cardboard box\nB. basket\nC. wall lamp\nD. blanket", "choices": ["A. cardboard box", "B. basket", "C. wall lamp", "D. blanket"], "answer": "C. wall lamp", "target": "backpack", "split": "train"}
{"id": "7dab70c8c8_2", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the wall lamp, about 11 cm away. It is also to the side of and below the cardboard box, about 32 cm away.", "target": "backpack", "split": "train"}
{"id": "7dab70c8c8_3", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"wall lamp\", \"references\": [{\"object\": \"wall lamp\", \"relation\": \"side_below\", \"distance_cm\": 10.6}, {\"object\": \"cardboard box\", \"relation\": \"side_below\", \"distance_cm\": 32.2}]}", "target": "backpack", "split": "train"}
{"id": "7dab70c8c8_4", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "7dab70c8c8_5", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. blanket\nB. basket\nC. speaker\nD. tissue box", "choices": ["A. blanket", "B. basket", "C. speaker", "D. tissue box"], "answer": "D. tissue box", "target": "bottle", "split": "train"}
{"id": "7dab70c8c8_6", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the desk. It is also next to the tissue box, about 9 cm away. It is further near the speaker, about 16 cm away.", "target": "bottle", "split": "train"}
{"id": "7dab70c8c8_7", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"tissue box\", \"relation\": \"next_to\", \"distance_cm\": 8.7}, {\"object\": \"speaker\", \"relation\": \"near\", \"distance_cm\": 15.8}]}", "target": "bottle", "split": "train"}
{"id": "7dab70c8c8_8", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "7dab70c8c8_9", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. office chair\nB. pedestal fan\nC. desk\nD. cooking pot", "choices": ["A. office chair", "B. pedestal fan", "C. desk", "D. cooking pot"], "answer": "B. pedestal fan", "target": "extension cord", "split": "train"}
{"id": "7dab70c8c8_10", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the pedestal fan, about 16 cm away. It is also to the side of and below the office chair, about 32 cm away. It is further to the side of and below the desk, about 40 cm away.", "target": "extension cord", "split": "train"}
{"id": "7dab70c8c8_11", "scene_id": "7dab70c8c8", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"pedestal fan\", \"relation\": \"side_below\", \"distance_cm\": 16.1}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 32.4}]}", "target": "extension cord", "split": "train"}
{"id": "7e09430da7_0", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "7e09430da7_1", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cooking pot\nB. split\nC. cardboard box\nD. tote bag", "choices": ["A. cooking pot", "B. split", "C. cardboard box", "D. tote bag"], "answer": "D. tote bag", "target": "backpack", "split": "train"}
{"id": "7e09430da7_2", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the tote bag, about 10 cm away. It is also to the side of and below the split, about 20 cm away. It is further to the side of and above the cardboard box, about 25 cm away.", "target": "backpack", "split": "train"}
{"id": "7e09430da7_3", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"tote bag\", \"relation\": \"side_above\", \"distance_cm\": 10.1}, {\"object\": \"split\", \"relation\": \"side_below\", \"distance_cm\": 20.4}]}", "target": "backpack", "split": "train"}
{"id": "7e09430da7_4", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "7e09430da7_5", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. coffee table\nC. pillow\nD. toy", "choices": ["A. monitor", "B. coffee table", "C. pillow", "D. toy"], "answer": "C. pillow", "target": "bottle", "split": "train"}
{"id": "7e09430da7_6", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the pillow, about 10 cm away. It is also near the toy, about 28 cm away. It is further to the side of and above the coffee table, about 39 cm away.", "target": "bottle", "split": "train"}
{"id": "7e09430da7_7", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"computer table\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 9.6}, {\"object\": \"toy\", \"relation\": \"near\", \"distance_cm\": 28.2}]}", "target": "bottle", "split": "train"}
{"id": "7e09430da7_8", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "7e09430da7_9", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. glass\nB. box\nC. glass plate\nD. mug", "choices": ["A. glass", "B. box", "C. glass plate", "D. mug"], "answer": "D. mug", "target": "cup", "split": "train"}
{"id": "7e09430da7_10", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the mug. It is also to the side of and above the box, about 6 cm away. It is further near the glass, about 16 cm away.", "target": "cup", "split": "train"}
{"id": "7e09430da7_11", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 2.0}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 6.4}]}", "target": "cup", "split": "train"}
{"id": "7e09430da7_12", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "No", "target": "notebook", "split": "train"}
{"id": "7e09430da7_13", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. monitor\nB. pillow\nC. coffee table\nD. toy", "choices": ["A. monitor", "B. pillow", "C. coffee table", "D. toy"], "answer": "B. pillow", "target": "spray bottle", "split": "train"}
{"id": "7e09430da7_14", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and above the pillow, about 16 cm away. It is also near the toy, about 22 cm away. It is further to the side of and above the coffee table, about 34 cm away.", "target": "spray bottle", "split": "train"}
{"id": "7e09430da7_15", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"computer table\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 16.1}, {\"object\": \"toy\", \"relation\": \"near\", \"distance_cm\": 22.1}]}", "target": "spray bottle", "split": "train"}
{"id": "7e09430da7_16", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "7e09430da7_17", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. cardboard box\nB. cooking pot\nC. split\nD. backpack", "choices": ["A. cardboard box", "B. cooking pot", "C. split", "D. backpack"], "answer": "D. backpack", "target": "tote bag", "split": "train"}
{"id": "7e09430da7_18", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and below the backpack, about 10 cm away. It is also to the side of and below the door, about 16 cm away. It is further to the side of and above the cooking pot, about 23 cm away.", "target": "tote bag", "split": "train"}
{"id": "7e09430da7_19", "scene_id": "7e09430da7", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"backpack\", \"relation\": \"side_below\", \"distance_cm\": 10.1}, {\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 16.0}]}", "target": "tote bag", "split": "train"}
{"id": "7e7d2e8640_0", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "7e7d2e8640_1", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "7e7d2e8640_2", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. chair\nB. backpack\nC. whiteboard marker\nD. heater", "choices": ["A. chair", "B. backpack", "C. whiteboard marker", "D. heater"], "answer": "A. chair", "target": "power strip", "split": "train"}
{"id": "7e7d2e8640_3", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the chair, about 32 cm away. It is also to the side of and below the heater, about 35 cm away. It is further to the side of and below the whiteboard marker, about 67 cm away.", "target": "power strip", "split": "train"}
{"id": "7e7d2e8640_4", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 31.6}, {\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 34.9}]}", "target": "power strip", "split": "train"}
{"id": "7e7d2e8640_5", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "7e7d2e8640_6", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. books\nB. backpack\nC. cabinet\nD. office chair", "choices": ["A. books", "B. backpack", "C. cabinet", "D. office chair"], "answer": "A. books", "target": "stapler", "split": "train"}
{"id": "7e7d2e8640_7", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is to the side of and below the books, about 20 cm away. It is also to the side of and above the office chair, about 22 cm away.", "target": "stapler", "split": "train"}
{"id": "7e7d2e8640_8", "scene_id": "7e7d2e8640", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 20.1}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 22.4}]}", "target": "stapler", "split": "train"}
{"id": "7eac902fd5_0", "scene_id": "7eac902fd5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "7eac902fd5_1", "scene_id": "7eac902fd5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. sink\nB. paper towel\nC. towel\nD. bottle", "choices": ["A. sink", "B. paper towel", "C. towel", "D. bottle"], "answer": "D. bottle", "target": "mug", "split": "train"}
{"id": "7eac902fd5_2", "scene_id": "7eac902fd5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the shelf. It is also to the side of and above the sink, about 29 cm away. It is further to the side of and above the towel, about 40 cm away.", "target": "mug", "split": "train"}
{"id": "7eac902fd5_3", "scene_id": "7eac902fd5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 28.7}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 40.1}]}", "target": "mug", "split": "train"}
{"id": "7ed0e10b52_0", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "7ed0e10b52_1", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. box\nB. dustpan\nC. power socket\nD. staircase railing", "choices": ["A. box", "B. dustpan", "C. power socket", "D. staircase railing"], "answer": "A. box", "target": "charger", "split": "train"}
{"id": "7ed0e10b52_2", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is above the box, about 4 cm away. It is also to the side of and above the dustpan, about 13 cm away. It is further to the side of and below the staircase railing, about 14 cm away.", "target": "charger", "split": "train"}
{"id": "7ed0e10b52_3", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"box\", \"references\": [{\"object\": \"box\", \"relation\": \"above\", \"distance_cm\": 3.7}, {\"object\": \"dustpan\", \"relation\": \"side_above\", \"distance_cm\": 13.2}]}", "target": "charger", "split": "train"}
{"id": "7ed0e10b52_4", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "7ed0e10b52_5", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. staircase railing\nB. dustpan\nC. laptop charger\nD. power socket", "choices": ["A. staircase railing", "B. dustpan", "C. laptop charger", "D. power socket"], "answer": "B. dustpan", "target": "extension cord", "split": "train"}
{"id": "7ed0e10b52_6", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and above the dustpan, about 9 cm away. It is also to the side of and below the staircase railing, about 11 cm away. It is further to the side of and above the laptop charger, about 24 cm away.", "target": "extension cord", "split": "train"}
{"id": "7ed0e10b52_7", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"box\", \"references\": [{\"object\": \"dustpan\", \"relation\": \"side_above\", \"distance_cm\": 9.1}, {\"object\": \"staircase railing\", \"relation\": \"side_below\", \"distance_cm\": 11.3}]}", "target": "extension cord", "split": "train"}
{"id": "7ed0e10b52_8", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "7ed0e10b52_9", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. router\nB. box\nC. dustpan\nD. reusable plastic bag", "choices": ["A. router", "B. box", "C. dustpan", "D. reusable plastic bag"], "answer": "D. reusable plastic bag", "target": "laptop charger", "split": "train"}
{"id": "7ed0e10b52_10", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the reusable plastic bag, about 10 cm away. It is also next to the box, about 15 cm away. It is further near the dustpan, about 18 cm away.", "target": "laptop charger", "split": "train"}
{"id": "7ed0e10b52_11", "scene_id": "7ed0e10b52", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"box\", \"references\": [{\"object\": \"reusable plastic bag\", \"relation\": \"side_below\", \"distance_cm\": 9.7}, {\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 14.7}]}", "target": "laptop charger", "split": "train"}
{"id": "7efa3f6b1b_0", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "7efa3f6b1b_1", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. heater\nC. mouse\nD. curtain", "choices": ["A. chair", "B. heater", "C. mouse", "D. curtain"], "answer": "B. heater", "target": "keyboard", "split": "train"}
{"id": "7efa3f6b1b_2", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the heater, about 6 cm away. It is further to the side of and below the curtain, about 8 cm away.", "target": "keyboard", "split": "train"}
{"id": "7efa3f6b1b_3", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 6.3}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 7.7}]}", "target": "keyboard", "split": "train"}
{"id": "7efa3f6b1b_4", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "7efa3f6b1b_5", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. box\nC. chair\nD. papers", "choices": ["A. keyboard", "B. box", "C. chair", "D. papers"], "answer": "B. box", "target": "mouse", "split": "train"}
{"id": "7efa3f6b1b_6", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the box, about 7 cm away. It is further next to the papers, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "7efa3f6b1b_7", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 7.4}, {\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 8.9}]}", "target": "mouse", "split": "train"}
{"id": "7efa3f6b1b_8", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "7efa3f6b1b_9", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also next to the laptop, about 7 cm away. It is further next to the box, about 7 cm away.", "target": "notebook", "split": "train"}
{"id": "7efa3f6b1b_10", "scene_id": "7efa3f6b1b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.6}, {\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "notebook", "split": "train"}
{"id": "7f22d5ef1b_0", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "7f22d5ef1b_1", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. paper\nC. food\nD. clothes", "choices": ["A. bag", "B. paper", "C. food", "D. clothes"], "answer": "D. clothes", "target": "backpack", "split": "train"}
{"id": "7f22d5ef1b_2", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the clothes. It is also next to the bag, about 4 cm away. It is further to the side of and below the window, about 18 cm away.", "target": "backpack", "split": "train"}
{"id": "7f22d5ef1b_3", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"window\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_above\", \"distance_cm\": 2.4}, {\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 3.6}]}", "target": "backpack", "split": "train"}
{"id": "7f22d5ef1b_4", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "7f22d5ef1b_5", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. bottle\nB. laptop\nC. bag\nD. laptop stand", "choices": ["A. bottle", "B. laptop", "C. bag", "D. laptop stand"], "answer": "B. laptop", "target": "keyboard", "split": "train"}
{"id": "7f22d5ef1b_6", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the laptop, about 3 cm away. It is further to the side of and below the bottle, about 7 cm away.", "target": "keyboard", "split": "train"}
{"id": "7f22d5ef1b_7", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"under\", \"distance_cm\": 3.2}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 6.7}]}", "target": "keyboard", "split": "train"}
{"id": "7f22d5ef1b_8", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "7f22d5ef1b_9", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. note\nB. monitor\nC. office chair\nD. laptop", "choices": ["A. note", "B. monitor", "C. office chair", "D. laptop"], "answer": "A. note", "target": "mouse", "split": "train"}
{"id": "7f22d5ef1b_10", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the note, about 6 cm away. It is further to the side of and below the monitor, about 16 cm away.", "target": "mouse", "split": "train"}
{"id": "7f22d5ef1b_11", "scene_id": "7f22d5ef1b", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"note\", \"relation\": \"next_to\", \"distance_cm\": 5.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 16.2}]}", "target": "mouse", "split": "train"}
{"id": "7f4d173c9c_0", "scene_id": "7f4d173c9c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "7f4d173c9c_1", "scene_id": "7f4d173c9c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. books\nB. chair\nC. speaker\nD. tv", "choices": ["A. books", "B. chair", "C. speaker", "D. tv"], "answer": "B. chair", "target": "tv remote", "split": "train"}
{"id": "7f4d173c9c_2", "scene_id": "7f4d173c9c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is next to the chair, about 13 cm away. It is also near the speaker, about 17 cm away. It is further to the side of and below the tv, about 26 cm away.", "target": "tv remote", "split": "train"}
{"id": "7f4d173c9c_3", "scene_id": "7f4d173c9c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"chair\", \"relation\": \"next_to\", \"distance_cm\": 12.8}, {\"object\": \"speaker\", \"relation\": \"near\", \"distance_cm\": 16.6}]}", "target": "tv remote", "split": "train"}
{"id": "7f97f24691_0", "scene_id": "7f97f24691", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "7f97f24691_1", "scene_id": "7f97f24691", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. stool\nB. book\nC. table\nD. chair", "choices": ["A. stool", "B. book", "C. table", "D. chair"], "answer": "D. chair", "target": "laptop charger", "split": "train"}
{"id": "7f97f24691_2", "scene_id": "7f97f24691", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the chair, about 32 cm away. It is also to the side of and below the table, about 34 cm away. It is further to the side of and below the stool, about 45 cm away.", "target": "laptop charger", "split": "train"}
{"id": "7f97f24691_3", "scene_id": "7f97f24691", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 32.1}, {\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 33.9}]}", "target": "laptop charger", "split": "train"}
{"id": "7f97f24691_4", "scene_id": "7f97f24691", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "Yes", "target": "remote control", "split": "train"}
{"id": "7f97f24691_5", "scene_id": "7f97f24691", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote control in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote control is next to the dvd player, about 4 cm away. It is also under the books, about 4 cm away. It is further to the side of and below the speaker, about 28 cm away.", "target": "remote control", "split": "train"}
{"id": "7f97f24691_6", "scene_id": "7f97f24691", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote control in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote control\", \"support_surface\": \"books\", \"references\": [{\"object\": \"dvd player\", \"relation\": \"next_to\", \"distance_cm\": 3.9}, {\"object\": \"books\", \"relation\": \"under\", \"distance_cm\": 4.4}]}", "target": "remote control", "split": "train"}
{"id": "7fb8ff20e9_0", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "7fb8ff20e9_1", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. laptop charger\nB. window frame\nC. phone charger\nD. pipe", "choices": ["A. laptop charger", "B. window frame", "C. phone charger", "D. pipe"], "answer": "A. laptop charger", "target": "extension cord", "split": "train"}
{"id": "7fb8ff20e9_2", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is next to the laptop charger, about 6 cm away. It is also to the side of and below the pipe, about 17 cm away. It is further to the side of and below the window frame, about 24 cm away.", "target": "extension cord", "split": "train"}
{"id": "7fb8ff20e9_3", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"window frame\", \"references\": [{\"object\": \"laptop charger\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"pipe\", \"relation\": \"side_below\", \"distance_cm\": 17.0}]}", "target": "extension cord", "split": "train"}
{"id": "7fb8ff20e9_4", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "train"}
{"id": "7fb8ff20e9_5", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the table, about 8 cm away. It is also to the side of and above the mousepad, about 9 cm away. It is further to the side of and above the mini pc, about 14 cm away.", "target": "headphones", "split": "train"}
{"id": "7fb8ff20e9_6", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 8.0}, {\"object\": \"mousepad\", \"relation\": \"side_above\", \"distance_cm\": 8.7}]}", "target": "headphones", "split": "train"}
{"id": "7fb8ff20e9_7", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "7fb8ff20e9_8", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. table lamp\nC. mousepad\nD. monitor", "choices": ["A. mouse", "B. table lamp", "C. mousepad", "D. monitor"], "answer": "C. mousepad", "target": "keyboard", "split": "train"}
{"id": "7fb8ff20e9_9", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mousepad. It is further near the mouse, about 21 cm away.", "target": "keyboard", "split": "train"}
{"id": "7fb8ff20e9_10", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 2.3}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 20.9}]}", "target": "keyboard", "split": "train"}
{"id": "7fb8ff20e9_11", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "7fb8ff20e9_12", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. pipe\nB. phone charger\nC. extension cord\nD. window frame", "choices": ["A. pipe", "B. phone charger", "C. extension cord", "D. window frame"], "answer": "C. extension cord", "target": "laptop charger", "split": "train"}
{"id": "7fb8ff20e9_13", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is next to the extension cord, about 6 cm away. It is also next to the phone charger, about 12 cm away. It is further to the side of and below the window frame, about 12 cm away.", "target": "laptop charger", "split": "train"}
{"id": "7fb8ff20e9_14", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"window frame\", \"references\": [{\"object\": \"extension cord\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"phone charger\", \"relation\": \"next_to\", \"distance_cm\": 11.7}]}", "target": "laptop charger", "split": "train"}
{"id": "7fb8ff20e9_15", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "7fb8ff20e9_16", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. headphones\nB. monitor\nC. keyboard\nD. table", "choices": ["A. headphones", "B. monitor", "C. keyboard", "D. table"], "answer": "D. table", "target": "mouse", "split": "train"}
{"id": "7fb8ff20e9_17", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the table, about 8 cm away. It is also to the side of and below the monitor, about 20 cm away. It is further near the keyboard, about 21 cm away.", "target": "mouse", "split": "train"}
{"id": "7fb8ff20e9_18", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 8.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 19.6}]}", "target": "mouse", "split": "train"}
{"id": "7fb8ff20e9_19", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "7fb8ff20e9_20", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. mini pc\nB. monitor\nC. headphones\nD. keyboard", "choices": ["A. mini pc", "B. monitor", "C. headphones", "D. keyboard"], "answer": "D. keyboard", "target": "mousepad", "split": "train"}
{"id": "7fb8ff20e9_21", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also next to the keyboard. It is further to the side of and below the monitor, about 6 cm away.", "target": "mousepad", "split": "train"}
{"id": "7fb8ff20e9_22", "scene_id": "7fb8ff20e9", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.1}]}", "target": "mousepad", "split": "train"}
{"id": "80ffca8a48_0", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "80ffca8a48_1", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. table\nC. mug\nD. chair", "choices": ["A. mouse", "B. table", "C. mug", "D. chair"], "answer": "D. chair", "target": "keyboard", "split": "train"}
{"id": "80ffca8a48_2", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the chair, about 44 cm away. It is also near the mouse, about 59 cm away.", "target": "keyboard", "split": "train"}
{"id": "80ffca8a48_3", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 44.0}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 59.5}]}", "target": "keyboard", "split": "train"}
{"id": "80ffca8a48_4", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "80ffca8a48_5", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. chair\nC. mug\nD. table", "choices": ["A. keyboard", "B. chair", "C. mug", "D. table"], "answer": "C. mug", "target": "mouse", "split": "train"}
{"id": "80ffca8a48_6", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also near the mug, about 41 cm away. It is further to the side of and above the chair, about 44 cm away.", "target": "mouse", "split": "train"}
{"id": "80ffca8a48_7", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mug\", \"relation\": \"near\", \"distance_cm\": 40.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 44.4}]}", "target": "mouse", "split": "train"}
{"id": "80ffca8a48_8", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "80ffca8a48_9", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. table\nB. chair\nC. mouse\nD. keyboard", "choices": ["A. table", "B. chair", "C. mouse", "D. keyboard"], "answer": "C. mouse", "target": "mug", "split": "train"}
{"id": "80ffca8a48_10", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also near the mouse, about 41 cm away. It is further to the side of and above the chair, about 54 cm away.", "target": "mug", "split": "train"}
{"id": "80ffca8a48_11", "scene_id": "80ffca8a48", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 40.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 54.0}]}", "target": "mug", "split": "train"}
{"id": "8133208cb6_0", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "Yes", "target": "scissors", "split": "train"}
{"id": "8133208cb6_1", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scissors in this scene?\nA. split\nB. photocopy machine\nC. mug\nD. outlet", "choices": ["A. split", "B. photocopy machine", "C. mug", "D. outlet"], "answer": "C. mug", "target": "scissors", "split": "train"}
{"id": "8133208cb6_2", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is to the side of and above the photocopy machine, about 10 cm away. It is also to the side of and above the split, about 14 cm away. It is further to the side of and above the outlet, about 15 cm away.", "target": "scissors", "split": "train"}
{"id": "8133208cb6_3", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"photocopy machine\", \"relation\": \"side_above\", \"distance_cm\": 10.2}, {\"object\": \"split\", \"relation\": \"side_above\", \"distance_cm\": 13.7}]}", "target": "scissors", "split": "train"}
{"id": "8133208cb6_4", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "8133208cb6_5", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. sink\nB. bucket\nC. sponge\nD. liquid soap", "choices": ["A. sink", "B. bucket", "C. sponge", "D. liquid soap"], "answer": "A. sink", "target": "spray bottle", "split": "train"}
{"id": "8133208cb6_6", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and above the sink, about 4 cm away. It is also to the side of and above the sponge, about 11 cm away. It is further to the side of and above the liquid soap, about 34 cm away.", "target": "spray bottle", "split": "train"}
{"id": "8133208cb6_7", "scene_id": "8133208cb6", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 3.7}, {\"object\": \"sponge\", \"relation\": \"side_above\", \"distance_cm\": 11.0}]}", "target": "spray bottle", "split": "train"}
{"id": "816e996553_0", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "816e996553_1", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. clothes\nB. board\nC. office chair\nD. books", "choices": ["A. clothes", "B. board", "C. office chair", "D. books"], "answer": "D. books", "target": "cable", "split": "train"}
{"id": "816e996553_2", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is on the table. It is also next to the books, about 3 cm away. It is further to the side of and above the board, about 23 cm away.", "target": "cable", "split": "train"}
{"id": "816e996553_3", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"table\", \"references\": [{\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 3.1}, {\"object\": \"board\", \"relation\": \"side_above\", \"distance_cm\": 22.6}]}", "target": "cable", "split": "train"}
{"id": "816e996553_4", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "816e996553_5", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. office chair\nB. monitor\nC. table\nD. mouse", "choices": ["A. office chair", "B. monitor", "C. table", "D. mouse"], "answer": "C. table", "target": "keyboard", "split": "train"}
{"id": "816e996553_6", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the table, about 3 cm away. It is also next to the mouse, about 8 cm away. It is further to the side of and below the monitor, about 16 cm away.", "target": "keyboard", "split": "train"}
{"id": "816e996553_7", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.7}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 8.0}]}", "target": "keyboard", "split": "train"}
{"id": "816e996553_8", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "816e996553_9", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. keyboard\nC. office chair\nD. speaker", "choices": ["A. monitor", "B. keyboard", "C. office chair", "D. speaker"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "816e996553_10", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 8 cm away. It is further to the side of and below the monitor, about 21 cm away.", "target": "mouse", "split": "train"}
{"id": "816e996553_11", "scene_id": "816e996553", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 21.3}]}", "target": "mouse", "split": "train"}
{"id": "824d9cfa6e_0", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "824d9cfa6e_1", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the notebook, about 5 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "bottle", "split": "train"}
{"id": "824d9cfa6e_2", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.5}]}", "target": "bottle", "split": "train"}
{"id": "824d9cfa6e_3", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "Yes", "target": "eraser", "split": "train"}
{"id": "824d9cfa6e_4", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the eraser in this scene?\nA. marker storage\nB. bin\nC. bottle\nD. bag", "choices": ["A. marker storage", "B. bin", "C. bottle", "D. bag"], "answer": "A. marker storage", "target": "eraser", "split": "train"}
{"id": "824d9cfa6e_5", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the eraser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The eraser is to the side of and below the marker storage, about 6 cm away.", "target": "eraser", "split": "train"}
{"id": "824d9cfa6e_6", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the eraser in structured JSON format.", "choices": null, "answer": "{\"target\": \"eraser\", \"support_surface\": \"marker storage\", \"references\": [{\"object\": \"marker storage\", \"relation\": \"side_below\", \"distance_cm\": 5.6}]}", "target": "eraser", "split": "train"}
{"id": "824d9cfa6e_7", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "824d9cfa6e_8", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. cup\nB. table\nC. bottle\nD. monitor", "choices": ["A. cup", "B. table", "C. bottle", "D. monitor"], "answer": "C. bottle", "target": "notebook", "split": "train"}
{"id": "824d9cfa6e_9", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also next to the bottle, about 5 cm away. It is further to the side of and below the monitor, about 9 cm away.", "target": "notebook", "split": "train"}
{"id": "824d9cfa6e_10", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.0}]}", "target": "notebook", "split": "train"}
{"id": "824d9cfa6e_11", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "824d9cfa6e_12", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. paper files\nB. cup\nC. plate\nD. table", "choices": ["A. paper files", "B. cup", "C. plate", "D. table"], "answer": "C. plate", "target": "pen", "split": "train"}
{"id": "824d9cfa6e_13", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the table. It is also next to the plate. It is further next to the paper files, about 13 cm away.", "target": "pen", "split": "train"}
{"id": "824d9cfa6e_14", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"plate\", \"relation\": \"next_to\", \"distance_cm\": 2.1}, {\"object\": \"paper files\", \"relation\": \"next_to\", \"distance_cm\": 12.8}]}", "target": "pen", "split": "train"}
{"id": "824d9cfa6e_15", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "824d9cfa6e_16", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil case is on the table. It is also next to the keyboard, about 8 cm away. It is further next to the plate, about 9 cm away.", "target": "pencil case", "split": "train"}
{"id": "824d9cfa6e_17", "scene_id": "824d9cfa6e", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.2}, {\"object\": \"plate\", \"relation\": \"next_to\", \"distance_cm\": 9.0}]}", "target": "pencil case", "split": "train"}
{"id": "82f448db76_0", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "82f448db76_1", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. drawer\nB. standing coat rack\nC. bag\nD. cabinet", "choices": ["A. drawer", "B. standing coat rack", "C. bag", "D. cabinet"], "answer": "A. drawer", "target": "backpack", "split": "train"}
{"id": "82f448db76_2", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the drawer, about 8 cm away. It is also to the side of and below the standing coat rack, about 14 cm away. It is further near the bag, about 23 cm away.", "target": "backpack", "split": "train"}
{"id": "82f448db76_3", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"drawer\", \"references\": [{\"object\": \"drawer\", \"relation\": \"side_below\", \"distance_cm\": 8.3}, {\"object\": \"standing coat rack\", \"relation\": \"side_below\", \"distance_cm\": 13.7}]}", "target": "backpack", "split": "train"}
{"id": "82f448db76_4", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "82f448db76_5", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. wall lamp\nB. light switch\nC. glass bottle\nD. thermos", "choices": ["A. wall lamp", "B. light switch", "C. glass bottle", "D. thermos"], "answer": "C. glass bottle", "target": "cup", "split": "train"}
{"id": "82f448db76_6", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the glass bottle, about 25 cm away. It is also to the side of and above the wall lamp, about 38 cm away. It is further near the window, about 49 cm away.", "target": "cup", "split": "train"}
{"id": "82f448db76_7", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"glass bottle\", \"relation\": \"side_below\", \"distance_cm\": 25.2}, {\"object\": \"wall lamp\", \"relation\": \"side_above\", \"distance_cm\": 38.0}]}", "target": "cup", "split": "train"}
{"id": "82f448db76_8", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "82f448db76_9", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. spectacle case\nB. chair\nC. thermos\nD. jar", "choices": ["A. spectacle case", "B. chair", "C. thermos", "D. jar"], "answer": "C. thermos", "target": "extension cord", "split": "train"}
{"id": "82f448db76_10", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the desk. It is also to the side of and below the window, about 12 cm away. It is further to the side of and below the thermos, about 13 cm away.", "target": "extension cord", "split": "train"}
{"id": "82f448db76_11", "scene_id": "82f448db76", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 12.0}, {\"object\": \"thermos\", \"relation\": \"side_below\", \"distance_cm\": 13.2}]}", "target": "extension cord", "split": "train"}
{"id": "84d71cada6_0", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "84d71cada6_1", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. electric toothbrush\nC. paper\nD. mouse", "choices": ["A. chair", "B. electric toothbrush", "C. paper", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "train"}
{"id": "84d71cada6_2", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the mouse, about 5 cm away. It is further to the side of and below the electric toothbrush, about 14 cm away.", "target": "keyboard", "split": "train"}
{"id": "84d71cada6_3", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"electric toothbrush\", \"relation\": \"side_below\", \"distance_cm\": 13.9}]}", "target": "keyboard", "split": "train"}
{"id": "84d71cada6_4", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "84d71cada6_5", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. desk\nB. chair\nC. keyboard\nD. paper", "choices": ["A. desk", "B. chair", "C. keyboard", "D. paper"], "answer": "A. desk", "target": "mouse", "split": "train"}
{"id": "84d71cada6_6", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the desk, about 3 cm away. It is also next to the keyboard, about 5 cm away. It is further next to the paper, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "84d71cada6_7", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 3.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.6}]}", "target": "mouse", "split": "train"}
{"id": "84d71cada6_8", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "84d71cada6_9", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. chair\nB. paper\nC. laptop stand\nD. monitor", "choices": ["A. chair", "B. paper", "C. laptop stand", "D. monitor"], "answer": "A. chair", "target": "mousepad", "split": "train"}
{"id": "84d71cada6_10", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the desk. It is also to the side of and above the chair, about 7 cm away. It is further to the side of and below the monitor, about 10 cm away.", "target": "mousepad", "split": "train"}
{"id": "84d71cada6_11", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 7.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.6}]}", "target": "mousepad", "split": "train"}
{"id": "84d71cada6_12", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "84d71cada6_13", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. book\nB. power sockets\nC. bookshelf\nD. router", "choices": ["A. book", "B. power sockets", "C. bookshelf", "D. router"], "answer": "D. router", "target": "mug", "split": "train"}
{"id": "84d71cada6_14", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and below the router, about 12 cm away. It is also to the side of and above the book, about 14 cm away. It is further to the side of and below the bookshelf, about 16 cm away.", "target": "mug", "split": "train"}
{"id": "84d71cada6_15", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"router\", \"relation\": \"side_below\", \"distance_cm\": 12.4}, {\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 14.1}]}", "target": "mug", "split": "train"}
{"id": "84d71cada6_16", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "84d71cada6_17", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the desk. It is also under the monitor, about 5 cm away. It is further to the side of and below the decoration, about 18 cm away.", "target": "pen", "split": "train"}
{"id": "84d71cada6_18", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.7}, {\"object\": \"decoration\", \"relation\": \"side_below\", \"distance_cm\": 17.6}]}", "target": "pen", "split": "train"}
{"id": "84d71cada6_19", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "84d71cada6_20", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. router\nB. foodstuffs\nC. doorframe\nD. bookshelf", "choices": ["A. router", "B. foodstuffs", "C. doorframe", "D. bookshelf"], "answer": "D. bookshelf", "target": "phone charger", "split": "train"}
{"id": "84d71cada6_21", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the bookshelf, about 23 cm away. It is also near the foodstuffs, about 26 cm away. It is further to the side of and above the router, about 26 cm away.", "target": "phone charger", "split": "train"}
{"id": "84d71cada6_22", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 23.0}, {\"object\": \"foodstuffs\", \"relation\": \"near\", \"distance_cm\": 25.7}]}", "target": "phone charger", "split": "train"}
{"id": "84d71cada6_23", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "Yes", "target": "scissors", "split": "train"}
{"id": "84d71cada6_24", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scissors in this scene?\nA. desk\nB. dustpan\nC. plastic food container\nD. cardboard box", "choices": ["A. desk", "B. dustpan", "C. plastic food container", "D. cardboard box"], "answer": "D. cardboard box", "target": "scissors", "split": "train"}
{"id": "84d71cada6_25", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is on the table. It is also to the side of and below the cardboard box, about 6 cm away. It is further to the side of and below the desk, about 8 cm away.", "target": "scissors", "split": "train"}
{"id": "84d71cada6_26", "scene_id": "84d71cada6", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cardboard box\", \"relation\": \"side_below\", \"distance_cm\": 6.0}, {\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 7.9}]}", "target": "scissors", "split": "train"}
{"id": "85dc2702b7_0", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "85dc2702b7_1", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glue bottle in this scene?\nA. notepad\nB. whiteboard markers\nC. chair\nD. monitor", "choices": ["A. notepad", "B. whiteboard markers", "C. chair", "D. monitor"], "answer": "B. whiteboard markers", "target": "glue bottle", "split": "train"}
{"id": "85dc2702b7_2", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glue bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glue bottle is on the table. It is also to the side of and above the whiteboard markers, about 11 cm away. It is further to the side of and below the monitor, about 18 cm away.", "target": "glue bottle", "split": "train"}
{"id": "85dc2702b7_3", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glue bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"glue bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"whiteboard markers\", \"relation\": \"side_above\", \"distance_cm\": 10.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.0}]}", "target": "glue bottle", "split": "train"}
{"id": "85dc2702b7_4", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "85dc2702b7_5", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. notebook\nC. laptop stand\nD. notepad", "choices": ["A. mouse", "B. notebook", "C. laptop stand", "D. notepad"], "answer": "B. notebook", "target": "keyboard", "split": "train"}
{"id": "85dc2702b7_6", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the notebook, about 3 cm away. It is further next to the mouse, about 7 cm away.", "target": "keyboard", "split": "train"}
{"id": "85dc2702b7_7", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 6.8}]}", "target": "keyboard", "split": "train"}
{"id": "85dc2702b7_8", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "85dc2702b7_9", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. notebook\nB. books\nC. chair\nD. keyboard", "choices": ["A. notebook", "B. books", "C. chair", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "85dc2702b7_10", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 7 cm away. It is also next to the notebook, about 14 cm away. It is further near the books, about 17 cm away.", "target": "mouse", "split": "train"}
{"id": "85dc2702b7_11", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"laptop stand\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 14.3}]}", "target": "mouse", "split": "train"}
{"id": "85dc2702b7_12", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "85dc2702b7_13", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. mouse\nB. keyboard\nC. laptop stand\nD. chair", "choices": ["A. mouse", "B. keyboard", "C. laptop stand", "D. chair"], "answer": "B. keyboard", "target": "notebook", "split": "train"}
{"id": "85dc2702b7_14", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also next to the keyboard, about 3 cm away. It is further to the side of and above the chair, about 4 cm away.", "target": "notebook", "split": "train"}
{"id": "85dc2702b7_15", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 4.4}]}", "target": "notebook", "split": "train"}
{"id": "85dc2702b7_16", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "85dc2702b7_17", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notepad in this scene?\nA. laptop stand\nB. notebook\nC. monitor\nD. keyboard", "choices": ["A. laptop stand", "B. notebook", "C. monitor", "D. keyboard"], "answer": "A. laptop stand", "target": "notepad", "split": "train"}
{"id": "85dc2702b7_18", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notepad is on the table. It is also to the side of and below the laptop stand, about 11 cm away. It is further near the keyboard, about 15 cm away.", "target": "notepad", "split": "train"}
{"id": "85dc2702b7_19", "scene_id": "85dc2702b7", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"notepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop stand\", \"relation\": \"side_below\", \"distance_cm\": 10.7}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.4}]}", "target": "notepad", "split": "train"}
{"id": "867d97cf3d_0", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "867d97cf3d_1", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the blanket, about 20 cm away. It is also to the side of and below the sofa pillow, about 20 cm away.", "target": "backpack", "split": "train"}
{"id": "867d97cf3d_2", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"sofa pillow\", \"references\": [{\"object\": \"blanket\", \"relation\": \"side_below\", \"distance_cm\": 20.0}, {\"object\": \"sofa pillow\", \"relation\": \"side_below\", \"distance_cm\": 20.3}]}", "target": "backpack", "split": "train"}
{"id": "867d97cf3d_3", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "867d97cf3d_4", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. pillow\nB. hairdryer attachment\nC. books\nD. hairdryer", "choices": ["A. pillow", "B. hairdryer attachment", "C. books", "D. hairdryer"], "answer": "C. books", "target": "extension cord", "split": "train"}
{"id": "867d97cf3d_5", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is next to the books, about 6 cm away. It is also to the side of and below the pillow, about 39 cm away. It is further to the side of and below the hairdryer, about 65 cm away.", "target": "extension cord", "split": "train"}
{"id": "867d97cf3d_6", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 39.1}]}", "target": "extension cord", "split": "train"}
{"id": "867d97cf3d_7", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "867d97cf3d_8", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. suitcase\nC. router\nD. sofa", "choices": ["A. chair", "B. suitcase", "C. router", "D. sofa"], "answer": "A. chair", "target": "mouse", "split": "train"}
{"id": "867d97cf3d_9", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is under the chair, about 4 cm away. It is also next to the router, about 7 cm away. It is further to the side of and above the suitcase, about 17 cm away.", "target": "mouse", "split": "train"}
{"id": "867d97cf3d_10", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"chair\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"router\", \"relation\": \"next_to\", \"distance_cm\": 7.5}]}", "target": "mouse", "split": "train"}
{"id": "867d97cf3d_11", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "867d97cf3d_12", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. office chair\nB. pencil holder\nC. heater\nD. decoration", "choices": ["A. office chair", "B. pencil holder", "C. heater", "D. decoration"], "answer": "B. pencil holder", "target": "notebook", "split": "train"}
{"id": "867d97cf3d_13", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and below the pencil holder, about 17 cm away. It is also to the side of and above the office chair, about 29 cm away. It is further to the side of and below the decoration, about 39 cm away.", "target": "notebook", "split": "train"}
{"id": "867d97cf3d_14", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"pencil holder\", \"relation\": \"side_below\", \"distance_cm\": 17.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 28.8}]}", "target": "notebook", "split": "train"}
{"id": "867d97cf3d_15", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "867d97cf3d_16", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. books\nB. jacket\nC. pencil holder\nD. desk", "choices": ["A. books", "B. jacket", "C. pencil holder", "D. desk"], "answer": "D. desk", "target": "pen", "split": "train"}
{"id": "867d97cf3d_17", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is next to the desk, about 4 cm away. It is also next to the books, about 6 cm away. It is further to the side of and below the pencil holder, about 23 cm away.", "target": "pen", "split": "train"}
{"id": "867d97cf3d_18", "scene_id": "867d97cf3d", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 5.7}]}", "target": "pen", "split": "train"}
{"id": "871efc90fa_0", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "871efc90fa_1", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "871efc90fa_2", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. bed\nB. pillow\nC. bedside table\nD. space heater", "choices": ["A. bed", "B. pillow", "C. bedside table", "D. space heater"], "answer": "C. bedside table", "target": "extension cord", "split": "train"}
{"id": "871efc90fa_3", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the bedside table, about 13 cm away. It is also to the side of and below the bed, about 14 cm away. It is further to the side of and below the space heater, about 17 cm away.", "target": "extension cord", "split": "train"}
{"id": "871efc90fa_4", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"bedside table\", \"relation\": \"side_below\", \"distance_cm\": 13.0}, {\"object\": \"bed\", \"relation\": \"side_below\", \"distance_cm\": 14.1}]}", "target": "extension cord", "split": "train"}
{"id": "871efc90fa_5", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "Yes", "target": "handbag", "split": "train"}
{"id": "871efc90fa_6", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the handbag in this scene?\nA. suitcase\nB. doorframe\nC. jacket\nD. clothes", "choices": ["A. suitcase", "B. doorframe", "C. jacket", "D. clothes"], "answer": "B. doorframe", "target": "handbag", "split": "train"}
{"id": "871efc90fa_7", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the handbag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The handbag is to the side of and above the door, about 10 cm away. It is also to the side of and above the doorframe, about 10 cm away. It is further to the side of and below the suitcase, about 26 cm away.", "target": "handbag", "split": "train"}
{"id": "871efc90fa_8", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the handbag in structured JSON format.", "choices": null, "answer": "{\"target\": \"handbag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"door\", \"relation\": \"side_above\", \"distance_cm\": 9.8}, {\"object\": \"doorframe\", \"relation\": \"side_above\", \"distance_cm\": 10.3}]}", "target": "handbag", "split": "train"}
{"id": "871efc90fa_9", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "871efc90fa_10", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. dress\nB. facemask\nC. clothes\nD. suitcase", "choices": ["A. dress", "B. facemask", "C. clothes", "D. suitcase"], "answer": "A. dress", "target": "tote bag", "split": "train"}
{"id": "871efc90fa_11", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is attached to the door. It is also to the side of and above the dress, about 11 cm away. It is further to the side of and below the suitcase, about 26 cm away.", "target": "tote bag", "split": "train"}
{"id": "871efc90fa_12", "scene_id": "871efc90fa", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"door\", \"references\": [{\"object\": \"dress\", \"relation\": \"side_above\", \"distance_cm\": 11.1}, {\"object\": \"suitcase\", \"relation\": \"side_below\", \"distance_cm\": 26.0}]}", "target": "tote bag", "split": "train"}
{"id": "872268d1bf_0", "scene_id": "872268d1bf", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "872268d1bf_1", "scene_id": "872268d1bf", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. cutting board\nB. kitchen counter\nC. kitchen utensil\nD. tap", "choices": ["A. cutting board", "B. kitchen counter", "C. kitchen utensil", "D. tap"], "answer": "A. cutting board", "target": "cup", "split": "train"}
{"id": "872268d1bf_2", "scene_id": "872268d1bf", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the cutting board, about 7 cm away. It is also next to the kitchen utensil, about 8 cm away. It is further near the kitchen counter, about 17 cm away.", "target": "cup", "split": "train"}
{"id": "872268d1bf_3", "scene_id": "872268d1bf", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"cutting board\", \"relation\": \"side_below\", \"distance_cm\": 6.7}, {\"object\": \"kitchen utensil\", \"relation\": \"next_to\", \"distance_cm\": 8.2}]}", "target": "cup", "split": "train"}
{"id": "87957b7c0b_0", "scene_id": "87957b7c0b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "88f265fe25_0", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "88f265fe25_1", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. computer tower\nB. telephone\nC. monitor\nD. office chair", "choices": ["A. computer tower", "B. telephone", "C. monitor", "D. office chair"], "answer": "D. office chair", "target": "mouse", "split": "train"}
{"id": "88f265fe25_2", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and above the office chair, about 17 cm away. It is further to the side of and below the monitor, about 24 cm away.", "target": "mouse", "split": "train"}
{"id": "88f265fe25_3", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 16.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 24.3}]}", "target": "mouse", "split": "train"}
{"id": "88f265fe25_4", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "88f265fe25_5", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. kettle\nB. sink\nC. whiteboard\nD. trash can", "choices": ["A. kettle", "B. sink", "C. whiteboard", "D. trash can"], "answer": "A. kettle", "target": "mug", "split": "train"}
{"id": "88f265fe25_6", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the cabinet. It is also to the side of and below the kettle, about 12 cm away. It is further to the side of and below the whiteboard, about 30 cm away.", "target": "mug", "split": "train"}
{"id": "88f265fe25_7", "scene_id": "88f265fe25", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"kettle\", \"relation\": \"side_below\", \"distance_cm\": 12.3}, {\"object\": \"whiteboard\", \"relation\": \"side_below\", \"distance_cm\": 30.2}]}", "target": "mug", "split": "train"}
{"id": "89214f3ca0_0", "scene_id": "89214f3ca0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "89214f3ca0_1", "scene_id": "89214f3ca0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the marker in this scene?\nA. plant\nB. cabel\nC. duster\nD. cpu", "choices": ["A. plant", "B. cabel", "C. duster", "D. cpu"], "answer": "C. duster", "target": "marker", "split": "train"}
{"id": "89214f3ca0_2", "scene_id": "89214f3ca0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the marker in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The marker is next to the duster, about 3 cm away. It is also to the side of and above the plant, about 47 cm away.", "target": "marker", "split": "train"}
{"id": "89214f3ca0_3", "scene_id": "89214f3ca0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the marker in structured JSON format.", "choices": null, "answer": "{\"target\": \"marker\", \"support_surface\": \"duster\", \"references\": [{\"object\": \"duster\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"plant\", \"relation\": \"side_above\", \"distance_cm\": 46.5}]}", "target": "marker", "split": "train"}
{"id": "894dbd41f1_0", "scene_id": "894dbd41f1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "894dbd41f1_1", "scene_id": "894dbd41f1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. pillow\nB. monitor\nC. picture\nD. corner bench", "choices": ["A. pillow", "B. monitor", "C. picture", "D. corner bench"], "answer": "C. picture", "target": "cup", "split": "train"}
{"id": "894dbd41f1_2", "scene_id": "894dbd41f1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the picture, about 6 cm away. It is also to the side of and above the corner bench, about 11 cm away. It is further to the side of and below the window, about 20 cm away.", "target": "cup", "split": "train"}
{"id": "894dbd41f1_3", "scene_id": "894dbd41f1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"corner bench\", \"references\": [{\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 5.7}, {\"object\": \"corner bench\", \"relation\": \"side_above\", \"distance_cm\": 10.6}]}", "target": "cup", "split": "train"}
{"id": "8a20d62ac0_0", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "No", "target": "notebook", "split": "train"}
{"id": "8a20d62ac0_1", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 11 cm away. It is also to the side of and above the wall strip, about 42 cm away.", "target": "backpack", "split": "train"}
{"id": "8a20d62ac0_2", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 10.8}, {\"object\": \"wall strip\", \"relation\": \"side_above\", \"distance_cm\": 41.6}]}", "target": "backpack", "split": "train"}
{"id": "8a20d62ac0_3", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "8a20d62ac0_4", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. book\nC. monitor\nD. chair", "choices": ["A. office chair", "B. book", "C. monitor", "D. chair"], "answer": "D. chair", "target": "mouse", "split": "train"}
{"id": "8a20d62ac0_5", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the chair, about 31 cm away. It is also to the side of and above the office chair, about 38 cm away. It is further near the book, about 39 cm away.", "target": "mouse", "split": "train"}
{"id": "8a20d62ac0_6", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 31.1}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 38.5}]}", "target": "mouse", "split": "train"}
{"id": "8a20d62ac0_7", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "8a20d62ac0_8", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and below the window, about 8 cm away. It is also to the side of and below the windowsill, about 12 cm away. It is further to the side of and above the office table, about 13 cm away.", "target": "mug", "split": "train"}
{"id": "8a20d62ac0_9", "scene_id": "8a20d62ac0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 8.1}, {\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 12.4}]}", "target": "mug", "split": "train"}
{"id": "8b2c0938d6_0", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "8b2c0938d6_1", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. box\nC. paper\nD. cabinet", "choices": ["A. monitor", "B. box", "C. paper", "D. cabinet"], "answer": "C. paper", "target": "bottle", "split": "train"}
{"id": "8b2c0938d6_2", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the desk. It is also to the side of and above the paper, about 4 cm away. It is further near the box, about 18 cm away.", "target": "bottle", "split": "train"}
{"id": "8b2c0938d6_3", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"box\", \"relation\": \"near\", \"distance_cm\": 17.9}]}", "target": "bottle", "split": "train"}
{"id": "8b2c0938d6_4", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "Yes", "target": "calculator", "split": "train"}
{"id": "8b2c0938d6_5", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the calculator in this scene?\nA. mouse pad\nB. mouse\nC. keyboard\nD. plant", "choices": ["A. mouse pad", "B. mouse", "C. keyboard", "D. plant"], "answer": "A. mouse pad", "target": "calculator", "split": "train"}
{"id": "8b2c0938d6_6", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the calculator in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The calculator is on the desk. It is also next to the mouse pad, about 4 cm away. It is further next to the mouse, about 11 cm away.", "target": "calculator", "split": "train"}
{"id": "8b2c0938d6_7", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the calculator in structured JSON format.", "choices": null, "answer": "{\"target\": \"calculator\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse pad\", \"relation\": \"next_to\", \"distance_cm\": 3.7}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 11.1}]}", "target": "calculator", "split": "train"}
{"id": "8b2c0938d6_8", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "No", "target": "soap", "split": "train"}
{"id": "8b2c0938d6_9", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphone case in this scene?\nA. plug\nB. notebooks\nC. outlet\nD. cabinet", "choices": ["A. plug", "B. notebooks", "C. outlet", "D. cabinet"], "answer": "D. cabinet", "target": "headphone case", "split": "train"}
{"id": "8b2c0938d6_10", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphone case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphone case is to the side of and above the cabinet, about 6 cm away. It is also to the side of and above the notebooks, about 11 cm away. It is further to the side of and below the plug, about 14 cm away.", "target": "headphone case", "split": "train"}
{"id": "8b2c0938d6_11", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphone case in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphone case\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 6.4}, {\"object\": \"notebooks\", \"relation\": \"side_above\", \"distance_cm\": 10.8}]}", "target": "headphone case", "split": "train"}
{"id": "8b2c0938d6_12", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "8b2c0938d6_13", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the marker in this scene?\nA. chair\nB. office chair\nC. paper\nD. whiteboard", "choices": ["A. chair", "B. office chair", "C. paper", "D. whiteboard"], "answer": "D. whiteboard", "target": "marker", "split": "train"}
{"id": "8b2c0938d6_14", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the marker in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The marker is under the whiteboard, about 3 cm away. It is also to the side of and above the chair, about 27 cm away. It is further to the side of and above the office chair, about 39 cm away.", "target": "marker", "split": "train"}
{"id": "8b2c0938d6_15", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the marker in structured JSON format.", "choices": null, "answer": "{\"target\": \"marker\", \"support_surface\": \"pen tray\", \"references\": [{\"object\": \"whiteboard\", \"relation\": \"under\", \"distance_cm\": 3.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 26.6}]}", "target": "marker", "split": "train"}
{"id": "8b2c0938d6_16", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "8b2c0938d6_17", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 4 cm away. It is also to the side of and above the desk, about 5 cm away. It is further next to the calculator, about 11 cm away.", "target": "mouse", "split": "train"}
{"id": "8b2c0938d6_18", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.2}, {\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 4.8}]}", "target": "mouse", "split": "train"}
{"id": "8b2c0938d6_19", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "Yes", "target": "mouse pad", "split": "train"}
{"id": "8b2c0938d6_20", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. paper\nB. calculator\nC. cabinet\nD. insulated coffee mug", "choices": ["A. paper", "B. calculator", "C. cabinet", "D. insulated coffee mug"], "answer": "B. calculator", "target": "mouse pad", "split": "train"}
{"id": "8b2c0938d6_21", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the desk. It is also next to the calculator, about 4 cm away. It is further to the side of and below the insulated coffee mug, about 7 cm away.", "target": "mouse pad", "split": "train"}
{"id": "8b2c0938d6_22", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"calculator\", \"relation\": \"next_to\", \"distance_cm\": 3.7}, {\"object\": \"insulated coffee mug\", \"relation\": \"side_below\", \"distance_cm\": 7.0}]}", "target": "mouse pad", "split": "train"}
{"id": "8b2c0938d6_23", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "No", "target": "game controller", "split": "train"}
{"id": "8b2c0938d6_24", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power cord in this scene?\nA. kettle\nB. slipper\nC. papers\nD. office chair", "choices": ["A. kettle", "B. slipper", "C. papers", "D. office chair"], "answer": "A. kettle", "target": "power cord", "split": "train"}
{"id": "8b2c0938d6_25", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power cord is to the side of and below the kettle, about 7 cm away. It is also to the side of and below the office chair, about 9 cm away. It is further near the slipper, about 44 cm away.", "target": "power cord", "split": "train"}
{"id": "8b2c0938d6_26", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"power cord\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"kettle\", \"relation\": \"side_below\", \"distance_cm\": 7.4}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 9.2}]}", "target": "power cord", "split": "train"}
{"id": "8b2c0938d6_27", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "8b2c0938d6_28", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. keyboard\nB. radiator\nC. window sill\nD. outlet", "choices": ["A. keyboard", "B. radiator", "C. window sill", "D. outlet"], "answer": "D. outlet", "target": "spray bottle", "split": "train"}
{"id": "8b2c0938d6_29", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the cabinet. It is also under the outlet. It is further under the window sill, about 4 cm away.", "target": "spray bottle", "split": "train"}
{"id": "8b2c0938d6_30", "scene_id": "8b2c0938d6", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"outlet\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"window sill\", \"relation\": \"under\", \"distance_cm\": 4.5}]}", "target": "spray bottle", "split": "train"}
{"id": "8bafc0c9d6_0", "scene_id": "8bafc0c9d6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "No", "target": "bottle", "split": "train"}
{"id": "8bafc0c9d6_1", "scene_id": "8bafc0c9d6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the faucet, about 7 cm away. It is also to the side of and above the toliet paper, about 11 cm away. It is further to the side of and below the towel, about 24 cm away.", "target": "cup", "split": "train"}
{"id": "8bafc0c9d6_2", "scene_id": "8bafc0c9d6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"door\", \"references\": [{\"object\": \"faucet\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"toliet paper\", \"relation\": \"side_above\", \"distance_cm\": 11.1}]}", "target": "cup", "split": "train"}
{"id": "8bafc0c9d6_3", "scene_id": "8bafc0c9d6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "Yes", "target": "toothbrush", "split": "train"}
{"id": "8bafc0c9d6_4", "scene_id": "8bafc0c9d6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. faucet\nB. toliet paper\nC. tube\nD. sink", "choices": ["A. faucet", "B. toliet paper", "C. tube", "D. sink"], "answer": "D. sink", "target": "toothbrush", "split": "train"}
{"id": "8bafc0c9d6_5", "scene_id": "8bafc0c9d6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is to the side of and above the sink, about 9 cm away. It is also to the side of and above the faucet, about 11 cm away. It is further to the side of and above the toliet paper, about 20 cm away.", "target": "toothbrush", "split": "train"}
{"id": "8bafc0c9d6_6", "scene_id": "8bafc0c9d6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"door\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 8.9}, {\"object\": \"faucet\", \"relation\": \"side_above\", \"distance_cm\": 10.9}]}", "target": "toothbrush", "split": "train"}
{"id": "8c3169b088_0", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "8c3169b088_1", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. towel\nB. electric toothbrush\nC. spray bottle\nD. sink", "choices": ["A. towel", "B. electric toothbrush", "C. spray bottle", "D. sink"], "answer": "C. spray bottle", "target": "bottle", "split": "train"}
{"id": "8c3169b088_2", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the cabinet. It is also next to the spray bottle, about 14 cm away. It is further near the electric toothbrush, about 27 cm away.", "target": "bottle", "split": "train"}
{"id": "8c3169b088_3", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"spray bottle\", \"relation\": \"next_to\", \"distance_cm\": 13.8}, {\"object\": \"electric toothbrush\", \"relation\": \"near\", \"distance_cm\": 26.9}]}", "target": "bottle", "split": "train"}
{"id": "8c3169b088_4", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "8c3169b088_5", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. bottle\nB. electric toothbrush\nC. toilet paper\nD. towel", "choices": ["A. bottle", "B. electric toothbrush", "C. toilet paper", "D. towel"], "answer": "B. electric toothbrush", "target": "spray bottle", "split": "train"}
{"id": "8c3169b088_6", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the cabinet. It is also next to the electric toothbrush, about 10 cm away. It is further next to the bottle, about 14 cm away.", "target": "spray bottle", "split": "train"}
{"id": "8c3169b088_7", "scene_id": "8c3169b088", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"electric toothbrush\", \"relation\": \"next_to\", \"distance_cm\": 10.3}, {\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 13.8}]}", "target": "spray bottle", "split": "train"}
{"id": "8d0f714398_0", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "8d0f714398_1", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the mouse, about 3 cm away. It is further under the pen holder, about 4 cm away.", "target": "keyboard", "split": "train"}
{"id": "8d0f714398_2", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"pen holder\", \"relation\": \"under\", \"distance_cm\": 3.9}]}", "target": "keyboard", "split": "train"}
{"id": "8d0f714398_3", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "No", "target": "handbag", "split": "train"}
{"id": "8d0f714398_4", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. pen holder\nC. keyboard\nD. cable", "choices": ["A. monitor", "B. pen holder", "C. keyboard", "D. cable"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "8d0f714398_5", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the keyboard, about 3 cm away. It is further under the monitor, about 5 cm away.", "target": "mouse", "split": "train"}
{"id": "8d0f714398_6", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.9}]}", "target": "mouse", "split": "train"}
{"id": "8d0f714398_7", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "8d0f714398_8", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the desk. It is also to the side of and above the laptop, about 4 cm away. It is further next to the keyboard, about 4 cm away.", "target": "pen holder", "split": "train"}
{"id": "8d0f714398_9", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 3.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.9}]}", "target": "pen holder", "split": "train"}
{"id": "8d0f714398_10", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "8d0f714398_11", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. cabinet\nB. cable\nC. books\nD. book", "choices": ["A. cabinet", "B. cable", "C. books", "D. book"], "answer": "D. book", "target": "power strip", "split": "train"}
{"id": "8d0f714398_12", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the desk. It is also next to the book, about 4 cm away. It is further near the books, about 28 cm away.", "target": "power strip", "split": "train"}
{"id": "8d0f714398_13", "scene_id": "8d0f714398", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 3.6}, {\"object\": \"books\", \"relation\": \"near\", \"distance_cm\": 27.8}]}", "target": "power strip", "split": "train"}
{"id": "8d563fc2cc_0", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "8d563fc2cc_1", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. office chair\nC. monitor\nD. laptop", "choices": ["A. mouse", "B. office chair", "C. monitor", "D. laptop"], "answer": "A. mouse", "target": "keyboard", "split": "train"}
{"id": "8d563fc2cc_2", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 5 cm away. It is further to the side of and below the laptop, about 9 cm away.", "target": "keyboard", "split": "train"}
{"id": "8d563fc2cc_3", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.5}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 8.8}]}", "target": "keyboard", "split": "train"}
{"id": "8d563fc2cc_4", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "8d563fc2cc_5", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. mouse\nB. box\nC. telephone\nD. monitor", "choices": ["A. mouse", "B. box", "C. telephone", "D. monitor"], "answer": "B. box", "target": "mug", "split": "train"}
{"id": "8d563fc2cc_6", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the box, about 4 cm away. It is further next to the mouse, about 11 cm away.", "target": "mug", "split": "train"}
{"id": "8d563fc2cc_7", "scene_id": "8d563fc2cc", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 3.6}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 11.3}]}", "target": "mug", "split": "train"}
{"id": "8d82d0dac0_0", "scene_id": "8d82d0dac0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "8d82d0dac0_1", "scene_id": "8d82d0dac0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. tupperware\nB. soap dispenser\nC. sink\nD. faucet", "choices": ["A. tupperware", "B. soap dispenser", "C. sink", "D. faucet"], "answer": "A. tupperware", "target": "mug", "split": "train"}
{"id": "8d82d0dac0_2", "scene_id": "8d82d0dac0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and above the sink, about 20 cm away. It is also to the side of and below the soap dispenser, about 27 cm away. It is further to the side of and below the faucet, about 32 cm away.", "target": "mug", "split": "train"}
{"id": "8d82d0dac0_3", "scene_id": "8d82d0dac0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"tray\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 19.7}, {\"object\": \"soap dispenser\", \"relation\": \"side_below\", \"distance_cm\": 27.3}]}", "target": "mug", "split": "train"}
{"id": "8e00ac7f59_0", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "8e00ac7f59_1", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. box\nB. bottle\nC. office chair\nD. cabinet", "choices": ["A. box", "B. bottle", "C. office chair", "D. cabinet"], "answer": "C. office chair", "target": "backpack", "split": "train"}
{"id": "8e00ac7f59_2", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the office chair, about 41 cm away.", "target": "backpack", "split": "train"}
{"id": "8e00ac7f59_3", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"folding chair\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 40.6}]}", "target": "backpack", "split": "train"}
{"id": "8e00ac7f59_4", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "8e00ac7f59_5", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. keyboard\nB. split\nC. file tray\nD. bottle", "choices": ["A. keyboard", "B. split", "C. file tray", "D. bottle"], "answer": "D. bottle", "target": "mug", "split": "train"}
{"id": "8e00ac7f59_6", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the bottle, about 11 cm away. It is further near the keyboard, about 17 cm away.", "target": "mug", "split": "train"}
{"id": "8e00ac7f59_7", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 10.7}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 17.0}]}", "target": "mug", "split": "train"}
{"id": "8e00ac7f59_8", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "8e00ac7f59_9", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor base\nB. paper\nC. power socket\nD. table", "choices": ["A. monitor base", "B. paper", "C. power socket", "D. table"], "answer": "D. table", "target": "power strip", "split": "train"}
{"id": "8e00ac7f59_10", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the table, about 6 cm away. It is also to the side of and below the power socket, about 30 cm away. It is further to the side of and below the monitor base, about 76 cm away.", "target": "power strip", "split": "train"}
{"id": "8e00ac7f59_11", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 5.6}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 29.8}]}", "target": "power strip", "split": "train"}
{"id": "8e00ac7f59_12", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "8e00ac7f59_13", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also next to the box, about 5 cm away. It is further to the side of and below the bottle, about 6 cm away.", "target": "tissue box", "split": "train"}
{"id": "8e00ac7f59_14", "scene_id": "8e00ac7f59", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 4.7}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 6.3}]}", "target": "tissue box", "split": "train"}
{"id": "8e6ff28354_0", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "8e6ff28354_1", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. sofa\nB. whiteboard\nC. whiteboard eraser\nD. table", "choices": ["A. sofa", "B. whiteboard", "C. whiteboard eraser", "D. table"], "answer": "B. whiteboard", "target": "bottle", "split": "train"}
{"id": "8e6ff28354_2", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the whiteboard. It is also to the side of and below the whiteboard eraser, about 10 cm away.", "target": "bottle", "split": "train"}
{"id": "8e6ff28354_3", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"whiteboard stand\", \"references\": [{\"object\": \"whiteboard\", \"relation\": \"under\", \"distance_cm\": 2.3}, {\"object\": \"whiteboard eraser\", \"relation\": \"side_below\", \"distance_cm\": 10.0}]}", "target": "bottle", "split": "train"}
{"id": "8e6ff28354_4", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "Yes", "target": "tv remote", "split": "train"}
{"id": "8e6ff28354_5", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. sofa\nB. table\nC. whiteboard\nD. bottle", "choices": ["A. sofa", "B. table", "C. whiteboard", "D. bottle"], "answer": "A. sofa", "target": "tv remote", "split": "train"}
{"id": "8e6ff28354_6", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the table. It is also near the sofa, about 47 cm away.", "target": "tv remote", "split": "train"}
{"id": "8e6ff28354_7", "scene_id": "8e6ff28354", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"table\", \"references\": [{\"object\": \"sofa\", \"relation\": \"near\", \"distance_cm\": 46.6}]}", "target": "tv remote", "split": "train"}
{"id": "8f82c394d6_0", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "8f82c394d6_1", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. sofa\nB. mirror\nC. water filter\nD. kettel", "choices": ["A. sofa", "B. mirror", "C. water filter", "D. kettel"], "answer": "C. water filter", "target": "mug", "split": "train"}
{"id": "8f82c394d6_2", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the nightstand. It is also under the water filter, about 4 cm away. It is further to the side of and below the kettel, about 11 cm away.", "target": "mug", "split": "train"}
{"id": "8f82c394d6_3", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"water filter\", \"relation\": \"under\", \"distance_cm\": 4.0}, {\"object\": \"kettel\", \"relation\": \"side_below\", \"distance_cm\": 10.9}]}", "target": "mug", "split": "train"}
{"id": "8f82c394d6_4", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "8f82c394d6_5", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. water filter\nB. mirror\nC. kettel\nD. cabinet", "choices": ["A. water filter", "B. mirror", "C. kettel", "D. cabinet"], "answer": "A. water filter", "target": "tissue box", "split": "train"}
{"id": "8f82c394d6_6", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the nightstand. It is also to the side of and below the water filter, about 5 cm away. It is further to the side of and below the kettel, about 13 cm away.", "target": "tissue box", "split": "train"}
{"id": "8f82c394d6_7", "scene_id": "8f82c394d6", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"water filter\", \"relation\": \"side_below\", \"distance_cm\": 5.2}, {\"object\": \"kettel\", \"relation\": \"side_below\", \"distance_cm\": 13.5}]}", "target": "tissue box", "split": "train"}
{"id": "9071e139d9_0", "scene_id": "9071e139d9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "train"}
{"id": "9071e139d9_1", "scene_id": "9071e139d9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. table\nB. sofa\nC. cushion\nD. chessboard", "choices": ["A. table", "B. sofa", "C. cushion", "D. chessboard"], "answer": "D. chessboard", "target": "mug", "split": "train"}
{"id": "9071e139d9_2", "scene_id": "9071e139d9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also near the chessboard, about 27 cm away. It is further near the sofa, about 33 cm away.", "target": "mug", "split": "train"}
{"id": "9071e139d9_3", "scene_id": "9071e139d9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chessboard\", \"relation\": \"near\", \"distance_cm\": 27.3}, {\"object\": \"sofa\", \"relation\": \"near\", \"distance_cm\": 32.6}]}", "target": "mug", "split": "train"}
{"id": "9084d4cd97_0", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "9084d4cd97_1", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the paper, about 4 cm away. It is also under the box, about 4 cm away. It is further next to the mouse mat, about 7 cm away.", "target": "keyboard", "split": "train"}
{"id": "9084d4cd97_2", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 3.9}, {\"object\": \"box\", \"relation\": \"under\", \"distance_cm\": 4.5}]}", "target": "keyboard", "split": "train"}
{"id": "9084d4cd97_3", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "9084d4cd97_4", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office table\nB. keyboard\nC. box\nD. laptop", "choices": ["A. office table", "B. keyboard", "C. box", "D. laptop"], "answer": "A. office table", "target": "mouse", "split": "train"}
{"id": "9084d4cd97_5", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the office table, about 3 cm away. It is also near the keyboard, about 15 cm away. It is further to the side of and below the box, about 18 cm away.", "target": "mouse", "split": "train"}
{"id": "9084d4cd97_6", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 2.9}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.2}]}", "target": "mouse", "split": "train"}
{"id": "9084d4cd97_7", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "9084d4cd97_8", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. file box\nB. box\nC. paper\nD. folder", "choices": ["A. file box", "B. box", "C. paper", "D. folder"], "answer": "D. folder", "target": "pen holder", "split": "train"}
{"id": "9084d4cd97_9", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is to the side of and above the folder, about 6 cm away. It is also to the side of and above the paper, about 15 cm away. It is further to the side of and above the file box, about 18 cm away.", "target": "pen holder", "split": "train"}
{"id": "9084d4cd97_10", "scene_id": "9084d4cd97", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"folder\", \"relation\": \"side_above\", \"distance_cm\": 6.0}, {\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 15.5}]}", "target": "pen holder", "split": "train"}
{"id": "90d8e25cee_0", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "90d8e25cee_1", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. phone charger\nB. power outlets\nC. shelf rail\nD. duffel bag", "choices": ["A. phone charger", "B. power outlets", "C. shelf rail", "D. duffel bag"], "answer": "C. shelf rail", "target": "backpack", "split": "train"}
{"id": "90d8e25cee_2", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the shelf rail, about 11 cm away. It is also near the phone charger, about 43 cm away. It is further near the power outlets, about 44 cm away.", "target": "backpack", "split": "train"}
{"id": "90d8e25cee_3", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"shelf rail\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 11.1}, {\"object\": \"phone charger\", \"relation\": \"near\", \"distance_cm\": 43.1}]}", "target": "backpack", "split": "train"}
{"id": "90d8e25cee_4", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "90d8e25cee_5", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. shelf\nB. shelf rail\nC. cowboy hat\nD. deodorant bottle", "choices": ["A. shelf", "B. shelf rail", "C. cowboy hat", "D. deodorant bottle"], "answer": "A. shelf", "target": "bottle", "split": "train"}
{"id": "90d8e25cee_6", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the shelf. It is also to the side of and above the shelf rail, about 12 cm away. It is further to the side of and above the deodorant bottle, about 70 cm away.", "target": "bottle", "split": "train"}
{"id": "90d8e25cee_7", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf rail\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 2.3}, {\"object\": \"shelf rail\", \"relation\": \"side_above\", \"distance_cm\": 11.6}]}", "target": "bottle", "split": "train"}
{"id": "90d8e25cee_8", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the deodorant bottle appear?", "choices": null, "answer": "Yes", "target": "deodorant bottle", "split": "train"}
{"id": "90d8e25cee_9", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the deodorant bottle in this scene?\nA. digital piano\nB. jar\nC. cowboy hat\nD. shelf rail", "choices": ["A. digital piano", "B. jar", "C. cowboy hat", "D. shelf rail"], "answer": "A. digital piano", "target": "deodorant bottle", "split": "train"}
{"id": "90d8e25cee_10", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the deodorant bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The deodorant bottle is on the shelf. It is also to the side of and above the digital piano, about 19 cm away. It is further to the side of and below the shelf rail, about 22 cm away.", "target": "deodorant bottle", "split": "train"}
{"id": "90d8e25cee_11", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the deodorant bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"deodorant bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"digital piano\", \"relation\": \"side_above\", \"distance_cm\": 19.1}, {\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 21.7}]}", "target": "deodorant bottle", "split": "train"}
{"id": "90d8e25cee_12", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "90d8e25cee_13", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. thermos\nB. tumbler\nC. book\nD. monitor", "choices": ["A. thermos", "B. tumbler", "C. book", "D. monitor"], "answer": "C. book", "target": "extension cord", "split": "train"}
{"id": "90d8e25cee_14", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the desk. It is also next to the book, about 3 cm away. It is further to the side of and below the tumbler, about 11 cm away.", "target": "extension cord", "split": "train"}
{"id": "90d8e25cee_15", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"tumbler\", \"relation\": \"side_below\", \"distance_cm\": 11.2}]}", "target": "extension cord", "split": "train"}
{"id": "90d8e25cee_16", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "Yes", "target": "hairbrush", "split": "train"}
{"id": "90d8e25cee_17", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hairbrush in this scene?\nA. book\nB. tumbler\nC. monitor\nD. phone charger", "choices": ["A. book", "B. tumbler", "C. monitor", "D. phone charger"], "answer": "D. phone charger", "target": "hairbrush", "split": "train"}
{"id": "90d8e25cee_18", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hairbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hairbrush is on the desk. It is also next to the phone charger, about 7 cm away. It is further next to the book, about 13 cm away.", "target": "hairbrush", "split": "train"}
{"id": "90d8e25cee_19", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hairbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"hairbrush\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"phone charger\", \"relation\": \"next_to\", \"distance_cm\": 7.4}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 12.6}]}", "target": "hairbrush", "split": "train"}
{"id": "90d8e25cee_20", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "90d8e25cee_21", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. desk\nB. trolley\nC. monitor\nD. tumbler", "choices": ["A. desk", "B. trolley", "C. monitor", "D. tumbler"], "answer": "B. trolley", "target": "laptop charger", "split": "train"}
{"id": "90d8e25cee_22", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and above the trolley, about 3 cm away. It is also to the side of and below the desk, about 36 cm away. It is further to the side of and below the monitor, about 37 cm away.", "target": "laptop charger", "split": "train"}
{"id": "90d8e25cee_23", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"trolley\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 35.6}]}", "target": "laptop charger", "split": "train"}
{"id": "90d8e25cee_24", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "90d8e25cee_25", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and below the desk, about 23 cm away. It is also to the side of and below the floor lamp, about 24 cm away. It is further to the side of and above the laptop, about 25 cm away.", "target": "notebook", "split": "train"}
{"id": "90d8e25cee_26", "scene_id": "90d8e25cee", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 23.1}, {\"object\": \"floor lamp\", \"relation\": \"side_below\", \"distance_cm\": 23.9}]}", "target": "notebook", "split": "train"}
{"id": "91fc568d84_0", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "91fc568d84_1", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse, about 10 cm away. It is also to the side of and above the table, about 15 cm away. It is further to the side of and below the monitor stand, about 16 cm away.", "target": "keyboard", "split": "train"}
{"id": "91fc568d84_2", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 9.8}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 15.3}]}", "target": "keyboard", "split": "train"}
{"id": "91fc568d84_3", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "91fc568d84_4", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. docking station\nC. monitor stand\nD. table", "choices": ["A. keyboard", "B. docking station", "C. monitor stand", "D. table"], "answer": "A. keyboard", "target": "mouse", "split": "train"}
{"id": "91fc568d84_5", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 10 cm away. It is also to the side of and below the monitor stand, about 15 cm away. It is further to the side of and above the table, about 18 cm away.", "target": "mouse", "split": "train"}
{"id": "91fc568d84_6", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.8}, {\"object\": \"monitor stand\", \"relation\": \"side_below\", \"distance_cm\": 14.7}]}", "target": "mouse", "split": "train"}
{"id": "91fc568d84_7", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "91fc568d84_8", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. monitor stand\nB. keyboard\nC. cabinet\nD. table", "choices": ["A. monitor stand", "B. keyboard", "C. cabinet", "D. table"], "answer": "D. table", "target": "mug", "split": "train"}
{"id": "91fc568d84_9", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and above the table. It is also to the side of and above the cabinet, about 16 cm away. It is further near the keyboard, about 25 cm away.", "target": "mug", "split": "train"}
{"id": "91fc568d84_10", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.0}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 16.2}]}", "target": "mug", "split": "train"}
{"id": "91fc568d84_11", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "91fc568d84_12", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. docking station\nB. mousepad\nC. mouse\nD. monitor", "choices": ["A. docking station", "B. mousepad", "C. mouse", "D. monitor"], "answer": "B. mousepad", "target": "pen holder", "split": "train"}
{"id": "91fc568d84_13", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also to the side of and above the mousepad, about 11 cm away. It is further to the side of and below the monitor, about 22 cm away.", "target": "pen holder", "split": "train"}
{"id": "91fc568d84_14", "scene_id": "91fc568d84", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"side_above\", \"distance_cm\": 10.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 21.5}]}", "target": "pen holder", "split": "train"}
{"id": "924b364b9f_0", "scene_id": "924b364b9f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "924b364b9f_1", "scene_id": "924b364b9f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the air fryer, about 9 cm away. It is also to the side of and below the coffee pot, about 9 cm away. It is further to the side of and below the coffee machine, about 15 cm away.", "target": "extension cord", "split": "train"}
{"id": "924b364b9f_2", "scene_id": "924b364b9f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"air fryer\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"coffee pot\", \"relation\": \"side_below\", \"distance_cm\": 8.8}]}", "target": "extension cord", "split": "train"}
{"id": "928c9da20c_0", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "928c9da20c_1", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the sponge, about 4 cm away. It is also next to the faucet, about 5 cm away. It is further to the side of and below the shower curtain, about 20 cm away.", "target": "bottle", "split": "train"}
{"id": "928c9da20c_2", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shower wall\", \"references\": [{\"object\": \"sponge\", \"relation\": \"side_above\", \"distance_cm\": 4.1}, {\"object\": \"faucet\", \"relation\": \"next_to\", \"distance_cm\": 5.0}]}", "target": "bottle", "split": "train"}
{"id": "928c9da20c_3", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "Yes", "target": "glasses case", "split": "train"}
{"id": "928c9da20c_4", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. office chair\nB. mug\nC. jacket\nD. jar", "choices": ["A. office chair", "B. mug", "C. jacket", "D. jar"], "answer": "B. mug", "target": "glasses case", "split": "train"}
{"id": "928c9da20c_5", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is on the table. It is also next to the mug, about 12 cm away. It is further to the side of and below the jar, about 18 cm away.", "target": "glasses case", "split": "train"}
{"id": "928c9da20c_6", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 11.7}, {\"object\": \"jar\", \"relation\": \"side_below\", \"distance_cm\": 18.1}]}", "target": "glasses case", "split": "train"}
{"id": "928c9da20c_7", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "928c9da20c_8", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the mousepad, about 7 cm away. It is further next to the laptop, about 7 cm away.", "target": "mouse", "split": "train"}
{"id": "928c9da20c_9", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 6.9}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.9}]}", "target": "mouse", "split": "train"}
{"id": "928c9da20c_10", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "928c9da20c_11", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. mouse\nB. mug\nC. laptop\nD. jar", "choices": ["A. mouse", "B. mug", "C. laptop", "D. jar"], "answer": "C. laptop", "target": "mousepad", "split": "train"}
{"id": "928c9da20c_12", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also next to the laptop, about 4 cm away. It is further next to the mouse, about 7 cm away.", "target": "mousepad", "split": "train"}
{"id": "928c9da20c_13", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 3.7}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 6.9}]}", "target": "mousepad", "split": "train"}
{"id": "928c9da20c_14", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "928c9da20c_15", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. mousepad\nB. glasses case\nC. jacket\nD. jar", "choices": ["A. mousepad", "B. glasses case", "C. jacket", "D. jar"], "answer": "D. jar", "target": "mug", "split": "train"}
{"id": "928c9da20c_16", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the jar, about 4 cm away. It is further next to the glasses case, about 12 cm away.", "target": "mug", "split": "train"}
{"id": "928c9da20c_17", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"jar\", \"relation\": \"next_to\", \"distance_cm\": 4.0}, {\"object\": \"glasses case\", \"relation\": \"next_to\", \"distance_cm\": 11.7}]}", "target": "mug", "split": "train"}
{"id": "928c9da20c_18", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "928c9da20c_19", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. sponge\nB. electrical panel\nC. kitchen towel\nD. bottle", "choices": ["A. sponge", "B. electrical panel", "C. kitchen towel", "D. bottle"], "answer": "D. bottle", "target": "spray bottle", "split": "train"}
{"id": "928c9da20c_20", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and below the electrical panel, about 31 cm away. It is also near the kitchen towel, about 34 cm away. It is further to the side of and above the sponge, about 47 cm away.", "target": "spray bottle", "split": "train"}
{"id": "928c9da20c_21", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"electrical panel\", \"relation\": \"side_below\", \"distance_cm\": 30.8}, {\"object\": \"kitchen towel\", \"relation\": \"near\", \"distance_cm\": 34.1}]}", "target": "spray bottle", "split": "train"}
{"id": "928c9da20c_22", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "928c9da20c_23", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. glasses case\nB. book\nC. table\nD. power socket", "choices": ["A. glasses case", "B. book", "C. table", "D. power socket"], "answer": "C. table", "target": "tote bag", "split": "train"}
{"id": "928c9da20c_24", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and above the table, about 6 cm away. It is also to the side of and below the book, about 31 cm away. It is further to the side of and above the power socket, about 34 cm away.", "target": "tote bag", "split": "train"}
{"id": "928c9da20c_25", "scene_id": "928c9da20c", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 30.8}]}", "target": "tote bag", "split": "train"}
{"id": "9444b90aaa_0", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "Yes", "target": "remote", "split": "train"}
{"id": "9444b90aaa_1", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. tape\nB. bottle\nC. spray bottle\nD. curtain", "choices": ["A. tape", "B. bottle", "C. spray bottle", "D. curtain"], "answer": "B. bottle", "target": "remote", "split": "train"}
{"id": "9444b90aaa_2", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is on the desk. It is also next to the bottle. It is further to the side of and below the curtain, about 6 cm away.", "target": "remote", "split": "train"}
{"id": "9444b90aaa_3", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 2.3}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 5.5}]}", "target": "remote", "split": "train"}
{"id": "9444b90aaa_4", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "No", "target": "phone charger", "split": "train"}
{"id": "9444b90aaa_5", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. bottle\nB. chair\nC. flowers\nD. tape", "choices": ["A. bottle", "B. chair", "C. flowers", "D. tape"], "answer": "A. bottle", "target": "spray bottle", "split": "train"}
{"id": "9444b90aaa_6", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the desk. It is also next to the bottle, about 15 cm away. It is further to the side of and above the tape, about 19 cm away.", "target": "spray bottle", "split": "train"}
{"id": "9444b90aaa_7", "scene_id": "9444b90aaa", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 15.0}, {\"object\": \"tape\", \"relation\": \"side_above\", \"distance_cm\": 18.9}]}", "target": "spray bottle", "split": "train"}
{"id": "94b1acde81_0", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "94b1acde81_1", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. table\nC. jacket\nD. chair", "choices": ["A. bag", "B. table", "C. jacket", "D. chair"], "answer": "B. table", "target": "backpack", "split": "train"}
{"id": "94b1acde81_2", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the table, about 4 cm away. It is also to the side of and above the jacket, about 14 cm away. It is further near the bag, about 19 cm away.", "target": "backpack", "split": "train"}
{"id": "94b1acde81_3", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 14.0}]}", "target": "backpack", "split": "train"}
{"id": "94b1acde81_4", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "94b1acde81_5", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. vacuum cleaner\nB. coat hanger\nC. jacket\nD. tote bag", "choices": ["A. vacuum cleaner", "B. coat hanger", "C. jacket", "D. tote bag"], "answer": "A. vacuum cleaner", "target": "bottle", "split": "train"}
{"id": "94b1acde81_6", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the vacuum cleaner, about 4 cm away. It is also to the side of and above the coat hanger, about 37 cm away. It is further to the side of and below the jacket, about 40 cm away.", "target": "bottle", "split": "train"}
{"id": "94b1acde81_7", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"suitcase\", \"references\": [{\"object\": \"vacuum cleaner\", \"relation\": \"side_above\", \"distance_cm\": 3.5}, {\"object\": \"coat hanger\", \"relation\": \"side_above\", \"distance_cm\": 37.4}]}", "target": "bottle", "split": "train"}
{"id": "94b1acde81_8", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "94b1acde81_9", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. cabinet\nB. dishcloth\nC. plastic wrap\nD. light switch", "choices": ["A. cabinet", "B. dishcloth", "C. plastic wrap", "D. light switch"], "answer": "A. cabinet", "target": "cup", "split": "train"}
{"id": "94b1acde81_10", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the cabinet, about 5 cm away. It is also to the side of and above the plastic wrap, about 7 cm away. It is further to the side of and above the dishcloth, about 12 cm away.", "target": "cup", "split": "train"}
{"id": "94b1acde81_11", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 5.1}, {\"object\": \"plastic wrap\", \"relation\": \"side_above\", \"distance_cm\": 6.8}]}", "target": "cup", "split": "train"}
{"id": "94b1acde81_12", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "94b1acde81_13", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. bottle\nB. power adapter\nC. card\nD. wet tissue", "choices": ["A. bottle", "B. power adapter", "C. card", "D. wet tissue"], "answer": "B. power adapter", "target": "extension cord", "split": "train"}
{"id": "94b1acde81_14", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is near the power adapter, about 15 cm away. It is also below the wet tissue, about 28 cm away. It is further to the side of and below the card, about 30 cm away.", "target": "extension cord", "split": "train"}
{"id": "94b1acde81_15", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"power adapter\", \"relation\": \"near\", \"distance_cm\": 15.4}, {\"object\": \"wet tissue\", \"relation\": \"below\", \"distance_cm\": 28.0}]}", "target": "extension cord", "split": "train"}
{"id": "94b1acde81_16", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "94b1acde81_17", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. power sockets\nB. card\nC. bottle\nD. cabinet", "choices": ["A. power sockets", "B. card", "C. bottle", "D. cabinet"], "answer": "D. cabinet", "target": "mug", "split": "train"}
{"id": "94b1acde81_18", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and above the cabinet, about 17 cm away. It is also to the side of and above the bottle, about 28 cm away. It is further to the side of and above the power sockets, about 43 cm away.", "target": "mug", "split": "train"}
{"id": "94b1acde81_19", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 17.0}, {\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 28.1}]}", "target": "mug", "split": "train"}
{"id": "94b1acde81_20", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "No", "target": "pen", "split": "train"}
{"id": "94b1acde81_21", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. laptop\nB. bottle\nC. chair\nD. pencil case", "choices": ["A. laptop", "B. bottle", "C. chair", "D. pencil case"], "answer": "D. pencil case", "target": "notebook", "split": "train"}
{"id": "94b1acde81_22", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also near the pencil case, about 20 cm away. It is further to the side of and above the chair, about 31 cm away.", "target": "notebook", "split": "train"}
{"id": "94b1acde81_23", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pencil case\", \"relation\": \"near\", \"distance_cm\": 19.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 31.4}]}", "target": "notebook", "split": "train"}
{"id": "94b1acde81_24", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "Yes", "target": "pencil case", "split": "train"}
{"id": "94b1acde81_25", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil case in this scene?\nA. laptop\nB. chair\nC. monitor\nD. notebook", "choices": ["A. laptop", "B. chair", "C. monitor", "D. notebook"], "answer": "D. notebook", "target": "pencil case", "split": "train"}
{"id": "94b1acde81_26", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil case is on the table. It is also near the notebook, about 20 cm away. It is further to the side of and below the monitor, about 26 cm away.", "target": "pencil case", "split": "train"}
{"id": "94b1acde81_27", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"notebook\", \"relation\": \"near\", \"distance_cm\": 19.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 26.1}]}", "target": "pencil case", "split": "train"}
{"id": "94b1acde81_28", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "94b1acde81_29", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. bottle\nB. extension cord\nC. phone charger\nD. card", "choices": ["A. bottle", "B. extension cord", "C. phone charger", "D. card"], "answer": "B. extension cord", "target": "power adapter", "split": "train"}
{"id": "94b1acde81_30", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is near the extension cord, about 15 cm away. It is also to the side of and below the phone charger, about 17 cm away. It is further below the bottle, about 32 cm away.", "target": "power adapter", "split": "train"}
{"id": "94b1acde81_31", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"extension cord\", \"relation\": \"near\", \"distance_cm\": 15.4}, {\"object\": \"phone charger\", \"relation\": \"side_below\", \"distance_cm\": 17.0}]}", "target": "power adapter", "split": "train"}
{"id": "94b1acde81_32", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "94b1acde81_33", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "33", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is under the curtain, about 3 cm away. It is also to the side of and above the suitcase, about 31 cm away. It is further to the side of and above the vacuum cleaner, about 44 cm away.", "target": "tote bag", "split": "train"}
{"id": "94b1acde81_34", "scene_id": "94b1acde81", "source_dataset": "scannetpp", "local_sample_id": "34", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"suitcase\", \"references\": [{\"object\": \"curtain\", \"relation\": \"under\", \"distance_cm\": 3.0}, {\"object\": \"suitcase\", \"relation\": \"side_above\", \"distance_cm\": 31.4}]}", "target": "tote bag", "split": "train"}
{"id": "94ee15e8ba_0", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "94ee15e8ba_1", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. pillar\nB. power socket\nC. light switch\nD. heater", "choices": ["A. pillar", "B. power socket", "C. light switch", "D. heater"], "answer": "C. light switch", "target": "bottle", "split": "train"}
{"id": "94ee15e8ba_2", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is above the light switch, about 9 cm away. It is also to the side of and below the pillar, about 24 cm away. It is further to the side of and above the heater, about 36 cm away.", "target": "bottle", "split": "train"}
{"id": "94ee15e8ba_3", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"box\", \"references\": [{\"object\": \"light switch\", \"relation\": \"above\", \"distance_cm\": 8.6}, {\"object\": \"pillar\", \"relation\": \"side_below\", \"distance_cm\": 23.6}]}", "target": "bottle", "split": "train"}
{"id": "94ee15e8ba_4", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "94ee15e8ba_5", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. keyboard\nB. split\nC. power socket\nD. office chair", "choices": ["A. keyboard", "B. split", "C. power socket", "D. office chair"], "answer": "A. keyboard", "target": "cable", "split": "train"}
{"id": "94ee15e8ba_6", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is on the table. It is also next to the keyboard, about 13 cm away. It is further to the side of and above the office chair, about 15 cm away.", "target": "cable", "split": "train"}
{"id": "94ee15e8ba_7", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 12.8}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 15.1}]}", "target": "cable", "split": "train"}
{"id": "94ee15e8ba_8", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "Yes", "target": "highlighter", "split": "train"}
{"id": "94ee15e8ba_9", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the highlighter in this scene?\nA. monitor\nB. pen\nC. laptop\nD. paper", "choices": ["A. monitor", "B. pen", "C. laptop", "D. paper"], "answer": "B. pen", "target": "highlighter", "split": "train"}
{"id": "94ee15e8ba_10", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the highlighter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The highlighter is on the table. It is also under the pen, about 3 cm away. It is further to the side of and below the monitor, about 23 cm away.", "target": "highlighter", "split": "train"}
{"id": "94ee15e8ba_11", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the highlighter in structured JSON format.", "choices": null, "answer": "{\"target\": \"highlighter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pen\", \"relation\": \"under\", \"distance_cm\": 3.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.4}]}", "target": "highlighter", "split": "train"}
{"id": "94ee15e8ba_12", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "94ee15e8ba_13", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. office chair\nC. cable\nD. cabinet", "choices": ["A. mouse", "B. office chair", "C. cable", "D. cabinet"], "answer": "C. cable", "target": "keyboard", "split": "train"}
{"id": "94ee15e8ba_14", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the cable, about 13 cm away. It is further near the mouse, about 16 cm away.", "target": "keyboard", "split": "train"}
{"id": "94ee15e8ba_15", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cable\", \"relation\": \"next_to\", \"distance_cm\": 12.8}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 15.8}]}", "target": "keyboard", "split": "train"}
{"id": "94ee15e8ba_16", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "94ee15e8ba_17", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. monitor\nB. table\nC. highlighter\nD. pencil", "choices": ["A. monitor", "B. table", "C. highlighter", "D. pencil"], "answer": "C. highlighter", "target": "pen", "split": "train"}
{"id": "94ee15e8ba_18", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is above the highlighter, about 3 cm away. It is also next to the pencil, about 7 cm away. It is further to the side of and above the table, about 10 cm away.", "target": "pen", "split": "train"}
{"id": "94ee15e8ba_19", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"highlighter\", \"relation\": \"above\", \"distance_cm\": 3.5}, {\"object\": \"pencil\", \"relation\": \"next_to\", \"distance_cm\": 6.7}]}", "target": "pen", "split": "train"}
{"id": "94ee15e8ba_20", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "94ee15e8ba_21", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also near the penholder, about 30 cm away. It is further to the side of and below the monitor, about 35 cm away.", "target": "pen holder", "split": "train"}
{"id": "94ee15e8ba_22", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"penholder\", \"relation\": \"near\", \"distance_cm\": 30.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 35.2}]}", "target": "pen holder", "split": "train"}
{"id": "94ee15e8ba_23", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "Yes", "target": "pencil", "split": "train"}
{"id": "94ee15e8ba_24", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil in this scene?\nA. pen\nB. table\nC. monitor\nD. paper", "choices": ["A. pen", "B. table", "C. monitor", "D. paper"], "answer": "B. table", "target": "pencil", "split": "train"}
{"id": "94ee15e8ba_25", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil is to the side of and above the table, about 3 cm away. It is also next to the pen, about 7 cm away. It is further to the side of and below the monitor, about 27 cm away.", "target": "pencil", "split": "train"}
{"id": "94ee15e8ba_26", "scene_id": "94ee15e8ba", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.5}, {\"object\": \"pen\", \"relation\": \"next_to\", \"distance_cm\": 6.7}]}", "target": "pencil", "split": "train"}
{"id": "95450f0c4e_0", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "95450f0c4e_1", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. papers\nB. laptop\nC. mouse\nD. book", "choices": ["A. papers", "B. laptop", "C. mouse", "D. book"], "answer": "B. laptop", "target": "keyboard", "split": "train"}
{"id": "95450f0c4e_2", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the laptop, about 7 cm away. It is further next to the mouse, about 9 cm away.", "target": "keyboard", "split": "train"}
{"id": "95450f0c4e_3", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 9.4}]}", "target": "keyboard", "split": "train"}
{"id": "95450f0c4e_4", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "95450f0c4e_5", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. laptop\nC. towel\nD. keyboard", "choices": ["A. chair", "B. laptop", "C. towel", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "95450f0c4e_6", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the keyboard, about 9 cm away. It is further to the side of and above the towel, about 21 cm away.", "target": "mouse", "split": "train"}
{"id": "95450f0c4e_7", "scene_id": "95450f0c4e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.4}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 20.9}]}", "target": "mouse", "split": "train"}
{"id": "956e415fb1_0", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "956e415fb1_1", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is below the sunglasses, about 39 cm away. It is also below the apple, about 39 cm away. It is further near the table, about 40 cm away.", "target": "backpack", "split": "train"}
{"id": "956e415fb1_2", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"sunglasses\", \"relation\": \"below\", \"distance_cm\": 38.9}, {\"object\": \"apple\", \"relation\": \"below\", \"distance_cm\": 39.4}]}", "target": "backpack", "split": "train"}
{"id": "956e415fb1_3", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "956e415fb1_4", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. sunglasses\nB. wardrobe\nC. jacket\nD. backpack", "choices": ["A. sunglasses", "B. wardrobe", "C. jacket", "D. backpack"], "answer": "A. sunglasses", "target": "headphones", "split": "train"}
{"id": "956e415fb1_5", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the sunglasses, about 8 cm away. It is also to the side of and below the jacket, about 11 cm away. It is further to the side of and below the wardrobe, about 24 cm away.", "target": "headphones", "split": "train"}
{"id": "956e415fb1_6", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"sunglasses\", \"relation\": \"next_to\", \"distance_cm\": 8.5}, {\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 11.4}]}", "target": "headphones", "split": "train"}
{"id": "956e415fb1_7", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lanyard appear?", "choices": null, "answer": "No", "target": "lanyard", "split": "train"}
{"id": "956e415fb1_8", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. bedframe\nB. blanket\nC. pillow\nD. bedside table", "choices": ["A. bedframe", "B. blanket", "C. pillow", "D. bedside table"], "answer": "C. pillow", "target": "tv remote", "split": "train"}
{"id": "956e415fb1_9", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is under the pillow, about 3 cm away. It is also to the side of and below the blanket, about 19 cm away. It is further to the side of and above the bedside table, about 60 cm away.", "target": "tv remote", "split": "train"}
{"id": "956e415fb1_10", "scene_id": "956e415fb1", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"pillow\", \"relation\": \"under\", \"distance_cm\": 3.1}, {\"object\": \"blanket\", \"relation\": \"side_below\", \"distance_cm\": 19.5}]}", "target": "tv remote", "split": "train"}
{"id": "95748dd597_0", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "95748dd597_1", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cleaning bucket\nB. cloth\nC. cabinet\nD. mop", "choices": ["A. cleaning bucket", "B. cloth", "C. cabinet", "D. mop"], "answer": "B. cloth", "target": "backpack", "split": "train"}
{"id": "95748dd597_2", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the cloth, about 3 cm away. It is also to the side of and above the cleaning bucket, about 8 cm away. It is further next to the mop, about 12 cm away.", "target": "backpack", "split": "train"}
{"id": "95748dd597_3", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cloth\", \"relation\": \"side_above\", \"distance_cm\": 3.3}, {\"object\": \"cleaning bucket\", \"relation\": \"side_above\", \"distance_cm\": 8.4}]}", "target": "backpack", "split": "train"}
{"id": "95748dd597_4", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "95748dd597_5", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. shelf rail\nB. bedside cabinet\nC. laptop\nD. stationary", "choices": ["A. shelf rail", "B. bedside cabinet", "C. laptop", "D. stationary"], "answer": "C. laptop", "target": "headphones", "split": "train"}
{"id": "95748dd597_6", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also near the laptop, about 17 cm away. It is further to the side of and below the shelf rail, about 21 cm away.", "target": "headphones", "split": "train"}
{"id": "95748dd597_7", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 17.2}, {\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 20.9}]}", "target": "headphones", "split": "train"}
{"id": "95748dd597_8", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "95748dd597_9", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. bag\nB. table\nC. laptop\nD. shelf rail", "choices": ["A. bag", "B. table", "C. laptop", "D. shelf rail"], "answer": "B. table", "target": "mouse", "split": "train"}
{"id": "95748dd597_10", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the table, about 4 cm away. It is also next to the laptop, about 15 cm away. It is further to the side of and below the shelf rail, about 52 cm away.", "target": "mouse", "split": "train"}
{"id": "95748dd597_11", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.5}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 14.8}]}", "target": "mouse", "split": "train"}
{"id": "95748dd597_12", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "95748dd597_13", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. laptop\nB. shelf rail\nC. box\nD. bag", "choices": ["A. laptop", "B. shelf rail", "C. box", "D. bag"], "answer": "A. laptop", "target": "mousepad", "split": "train"}
{"id": "95748dd597_14", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also next to the laptop, about 5 cm away. It is further to the side of and below the shelf rail, about 44 cm away.", "target": "mousepad", "split": "train"}
{"id": "95748dd597_15", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 5.4}, {\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 43.6}]}", "target": "mousepad", "split": "train"}
{"id": "95748dd597_16", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "95748dd597_17", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. books\nB. dumbbell\nC. bag\nD. pillow", "choices": ["A. books", "B. dumbbell", "C. bag", "D. pillow"], "answer": "C. bag", "target": "notebook", "split": "train"}
{"id": "95748dd597_18", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is above the bag, about 6 cm away. It is also to the side of and below the pillow, about 15 cm away. It is further near the dumbbell, about 19 cm away.", "target": "notebook", "split": "train"}
{"id": "95748dd597_19", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"bedside cabinet\", \"references\": [{\"object\": \"bag\", \"relation\": \"above\", \"distance_cm\": 6.1}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 14.9}]}", "target": "notebook", "split": "train"}
{"id": "95748dd597_20", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "95748dd597_21", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. books\nB. bag\nC. notebook\nD. dumbbell", "choices": ["A. books", "B. bag", "C. notebook", "D. dumbbell"], "answer": "A. books", "target": "power strip", "split": "train"}
{"id": "95748dd597_22", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the books, about 3 cm away. It is also to the side of and below the bag, about 24 cm away. It is further to the side of and below the dumbbell, about 30 cm away.", "target": "power strip", "split": "train"}
{"id": "95748dd597_23", "scene_id": "95748dd597", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bedside cabinet\", \"references\": [{\"object\": \"books\", \"relation\": \"under\", \"distance_cm\": 3.5}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 23.9}]}", "target": "power strip", "split": "train"}
{"id": "95b9971d01_0", "scene_id": "95b9971d01", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "95b9971d01_1", "scene_id": "95b9971d01", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. kitchen cabinet\nB. microwave\nC. kitchen towel\nD. dish soap", "choices": ["A. kitchen cabinet", "B. microwave", "C. kitchen towel", "D. dish soap"], "answer": "B. microwave", "target": "spray bottle", "split": "train"}
{"id": "95b9971d01_2", "scene_id": "95b9971d01", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and below the microwave, about 15 cm away. It is also near the dish soap, about 29 cm away. It is further to the side of and below the kitchen cabinet, about 33 cm away.", "target": "spray bottle", "split": "train"}
{"id": "95b9971d01_3", "scene_id": "95b9971d01", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"microwave\", \"relation\": \"side_below\", \"distance_cm\": 14.9}, {\"object\": \"dish soap\", \"relation\": \"near\", \"distance_cm\": 29.4}]}", "target": "spray bottle", "split": "train"}
{"id": "967d2a847a_0", "scene_id": "967d2a847a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "No", "target": "game controller", "split": "train"}
{"id": "967d2a847a_1", "scene_id": "967d2a847a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the kettle. It is further under the plates, about 3 cm away.", "target": "bottle", "split": "train"}
{"id": "967d2a847a_2", "scene_id": "967d2a847a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"kettle\", \"relation\": \"next_to\", \"distance_cm\": 2.3}, {\"object\": \"plates\", \"relation\": \"under\", \"distance_cm\": 3.0}]}", "target": "bottle", "split": "train"}
{"id": "967d2a847a_3", "scene_id": "967d2a847a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "967d2a847a_4", "scene_id": "967d2a847a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the jug, about 9 cm away. It is also to the side of and above the cabinet, about 9 cm away. It is further to the side of and above the kitchen towel, about 15 cm away.", "target": "cup", "split": "train"}
{"id": "967d2a847a_5", "scene_id": "967d2a847a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"jug\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 8.8}]}", "target": "cup", "split": "train"}
{"id": "979a852c94_0", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "Yes", "target": "adapter", "split": "train"}
{"id": "979a852c94_1", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the adapter in this scene?\nA. windowsill\nB. power socket\nC. chair\nD. cable", "choices": ["A. windowsill", "B. power socket", "C. chair", "D. cable"], "answer": "B. power socket", "target": "adapter", "split": "train"}
{"id": "979a852c94_2", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The adapter is to the side of and below the power socket, about 11 cm away. It is also to the side of and below the windowsill, about 43 cm away.", "target": "adapter", "split": "train"}
{"id": "979a852c94_3", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"adapter\", \"support_surface\": \"power strip\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 10.9}, {\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 43.0}]}", "target": "adapter", "split": "train"}
{"id": "979a852c94_4", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "979a852c94_5", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "979a852c94_6", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. windowsill\nB. cable\nC. power socket\nD. adapter", "choices": ["A. windowsill", "B. cable", "C. power socket", "D. adapter"], "answer": "C. power socket", "target": "power strip", "split": "train"}
{"id": "979a852c94_7", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the power socket, about 11 cm away. It is also to the side of and below the windowsill, about 48 cm away.", "target": "power strip", "split": "train"}
{"id": "979a852c94_8", "scene_id": "979a852c94", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"adapter\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 11.5}, {\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 48.3}]}", "target": "power strip", "split": "train"}
{"id": "9859de300f_0", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "9859de300f_1", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. kettle\nB. shelf\nC. coffee machine\nD. box", "choices": ["A. kettle", "B. shelf", "C. coffee machine", "D. box"], "answer": "D. box", "target": "cup", "split": "train"}
{"id": "9859de300f_2", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the shelf. It is also to the side of and below the box, about 16 cm away. It is further to the side of and above the coffee machine, about 22 cm away.", "target": "cup", "split": "train"}
{"id": "9859de300f_3", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 16.4}, {\"object\": \"coffee machine\", \"relation\": \"side_above\", \"distance_cm\": 22.2}]}", "target": "cup", "split": "train"}
{"id": "9859de300f_4", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "9859de300f_5", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. tray\nB. kitchen cabinet\nC. kitchen counter\nD. kettle", "choices": ["A. tray", "B. kitchen cabinet", "C. kitchen counter", "D. kettle"], "answer": "C. kitchen counter", "target": "mug", "split": "train"}
{"id": "9859de300f_6", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and above the kitchen counter, about 4 cm away. It is also to the side of and above the tray, about 22 cm away. It is further to the side of and below the kettle, about 27 cm away.", "target": "mug", "split": "train"}
{"id": "9859de300f_7", "scene_id": "9859de300f", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 4.2}, {\"object\": \"tray\", \"relation\": \"side_above\", \"distance_cm\": 22.1}]}", "target": "mug", "split": "train"}
{"id": "98cc0b7c9a_0", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "98cc0b7c9a_1", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bowls\nB. bottle\nC. air fryer\nD. power strip", "choices": ["A. bowls", "B. bottle", "C. air fryer", "D. power strip"], "answer": "D. power strip", "target": "backpack", "split": "train"}
{"id": "98cc0b7c9a_2", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the power strip, about 56 cm away.", "target": "backpack", "split": "train"}
{"id": "98cc0b7c9a_3", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"power strip\", \"relation\": \"side_above\", \"distance_cm\": 56.1}]}", "target": "backpack", "split": "train"}
{"id": "98cc0b7c9a_4", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "98cc0b7c9a_5", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. jacket\nB. monitor\nC. office chair\nD. microphone", "choices": ["A. jacket", "B. monitor", "C. office chair", "D. microphone"], "answer": "C. office chair", "target": "keyboard", "split": "train"}
{"id": "98cc0b7c9a_6", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the office chair, about 10 cm away. It is further to the side of and below the monitor, about 19 cm away.", "target": "keyboard", "split": "train"}
{"id": "98cc0b7c9a_7", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 10.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.9}]}", "target": "keyboard", "split": "train"}
{"id": "98cc0b7c9a_8", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "98cc0b7c9a_9", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. power socket\nB. box\nC. backpack\nD. chair", "choices": ["A. power socket", "B. box", "C. backpack", "D. chair"], "answer": "D. chair", "target": "power strip", "split": "train"}
{"id": "98cc0b7c9a_10", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the chair, about 15 cm away. It is also to the side of and below the box, about 32 cm away. It is further to the side of and below the power socket, about 54 cm away.", "target": "power strip", "split": "train"}
{"id": "98cc0b7c9a_11", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 14.8}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 32.4}]}", "target": "power strip", "split": "train"}
{"id": "98cc0b7c9a_12", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "98cc0b7c9a_13", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. bottle\nB. power socket\nC. bucket\nD. sink", "choices": ["A. bottle", "B. power socket", "C. bucket", "D. sink"], "answer": "C. bucket", "target": "spray bottle", "split": "train"}
{"id": "98cc0b7c9a_14", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is next to the bucket, about 10 cm away. It is also near the bottle, about 20 cm away. It is further to the side of and below the sink, about 22 cm away.", "target": "spray bottle", "split": "train"}
{"id": "98cc0b7c9a_15", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"bucket\", \"references\": [{\"object\": \"bucket\", \"relation\": \"next_to\", \"distance_cm\": 10.1}, {\"object\": \"bottle\", \"relation\": \"near\", \"distance_cm\": 19.5}]}", "target": "spray bottle", "split": "train"}
{"id": "98cc0b7c9a_16", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "98cc0b7c9a_17", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. backpack\nB. air fryer\nC. bottle\nD. door frame", "choices": ["A. backpack", "B. air fryer", "C. bottle", "D. door frame"], "answer": "D. door frame", "target": "tote bag", "split": "train"}
{"id": "98cc0b7c9a_18", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and above the door, about 13 cm away. It is also to the side of and below the door frame, about 15 cm away.", "target": "tote bag", "split": "train"}
{"id": "98cc0b7c9a_19", "scene_id": "98cc0b7c9a", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door\", \"relation\": \"side_above\", \"distance_cm\": 13.3}, {\"object\": \"door frame\", \"relation\": \"side_below\", \"distance_cm\": 15.0}]}", "target": "tote bag", "split": "train"}
{"id": "99010a8938_0", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "99010a8938_1", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. window frame\nB. office chair\nC. heater\nD. box", "choices": ["A. window frame", "B. office chair", "C. heater", "D. box"], "answer": "D. box", "target": "backpack", "split": "train"}
{"id": "99010a8938_2", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the box, about 7 cm away. It is also to the side of and above the office chair, about 16 cm away. It is further near the window frame, about 18 cm away.", "target": "backpack", "split": "train"}
{"id": "99010a8938_3", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 7.0}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 16.1}]}", "target": "backpack", "split": "train"}
{"id": "99010a8938_4", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "99010a8938_5", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. window lintel\nB. shelf rail\nC. pedestal fan\nD. bottle", "choices": ["A. window lintel", "B. shelf rail", "C. pedestal fan", "D. bottle"], "answer": "D. bottle", "target": "cup", "split": "train"}
{"id": "99010a8938_6", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is under the bottle. It is also to the side of and above the window lintel, about 4 cm away. It is further to the side of and below the pedestal fan, about 13 cm away.", "target": "cup", "split": "train"}
{"id": "99010a8938_7", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf rail\", \"references\": [{\"object\": \"bottle\", \"relation\": \"under\", \"distance_cm\": 2.5}, {\"object\": \"window lintel\", \"relation\": \"side_above\", \"distance_cm\": 4.0}]}", "target": "cup", "split": "train"}
{"id": "99010a8938_8", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "99010a8938_9", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. food bag\nB. luggage\nC. ball\nD. bag", "choices": ["A. food bag", "B. luggage", "C. ball", "D. bag"], "answer": "B. luggage", "target": "extension cord", "split": "train"}
{"id": "99010a8938_10", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the luggage, about 36 cm away. It is also to the side of and below the bag, about 71 cm away. It is further to the side of and below the food bag, about 79 cm away.", "target": "extension cord", "split": "train"}
{"id": "99010a8938_11", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"luggage\", \"relation\": \"side_below\", \"distance_cm\": 36.4}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 70.6}]}", "target": "extension cord", "split": "train"}
{"id": "99010a8938_12", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "99010a8938_13", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. curtain\nB. shelf\nC. office chair\nD. chair", "choices": ["A. curtain", "B. shelf", "C. office chair", "D. chair"], "answer": "A. curtain", "target": "keyboard", "split": "train"}
{"id": "99010a8938_14", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the curtain. It is further to the side of and below the shelf, about 14 cm away.", "target": "keyboard", "split": "train"}
{"id": "99010a8938_15", "scene_id": "99010a8938", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"curtain\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 14.3}]}", "target": "keyboard", "split": "train"}
{"id": "992cd74f2c_0", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "No", "target": "calculator", "split": "train"}
{"id": "992cd74f2c_1", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. power strip\nC. basket\nD. suitcase", "choices": ["A. bag", "B. power strip", "C. basket", "D. suitcase"], "answer": "D. suitcase", "target": "backpack", "split": "train"}
{"id": "992cd74f2c_2", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the suitcase, about 8 cm away. It is also to the side of and above the basket, about 50 cm away. It is further to the side of and above the power strip, about 73 cm away.", "target": "backpack", "split": "train"}
{"id": "992cd74f2c_3", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"suitcase\", \"relation\": \"side_below\", \"distance_cm\": 7.5}, {\"object\": \"basket\", \"relation\": \"side_above\", \"distance_cm\": 49.8}]}", "target": "backpack", "split": "train"}
{"id": "992cd74f2c_4", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "Yes", "target": "wallet", "split": "train"}
{"id": "992cd74f2c_5", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the wallet in this scene?\nA. bag\nB. shelf\nC. plant pot\nD. paper towel", "choices": ["A. bag", "B. shelf", "C. plant pot", "D. paper towel"], "answer": "D. paper towel", "target": "wallet", "split": "train"}
{"id": "992cd74f2c_6", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the wallet in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The wallet is to the side of and below the paper towel, about 14 cm away. It is also to the side of and below the bag, about 28 cm away. It is further to the side of and below the plant pot, about 29 cm away.", "target": "wallet", "split": "train"}
{"id": "992cd74f2c_7", "scene_id": "992cd74f2c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the wallet in structured JSON format.", "choices": null, "answer": "{\"target\": \"wallet\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"paper towel\", \"relation\": \"side_below\", \"distance_cm\": 13.7}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 28.3}]}", "target": "wallet", "split": "train"}
{"id": "99532296c7_0", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "99532296c7_1", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. coffee maker\nB. chair\nC. basket\nD. game console", "choices": ["A. coffee maker", "B. chair", "C. basket", "D. game console"], "answer": "A. coffee maker", "target": "bottle", "split": "train"}
{"id": "99532296c7_2", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and below the coffee maker, about 14 cm away. It is further to the side of and above the chair, about 24 cm away.", "target": "bottle", "split": "train"}
{"id": "99532296c7_3", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"coffee maker\", \"relation\": \"side_below\", \"distance_cm\": 14.4}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 23.7}]}", "target": "bottle", "split": "train"}
{"id": "99532296c7_4", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "99532296c7_5", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. power socket\nB. tv\nC. game controller\nD. game console", "choices": ["A. power socket", "B. tv", "C. game controller", "D. game console"], "answer": "D. game console", "target": "tv remote", "split": "train"}
{"id": "99532296c7_6", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is next to the game console, about 8 cm away. It is also to the side of and below the game controller, about 9 cm away. It is further to the side of and below the tv, about 29 cm away.", "target": "tv remote", "split": "train"}
{"id": "99532296c7_7", "scene_id": "99532296c7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"game console\", \"relation\": \"next_to\", \"distance_cm\": 7.6}, {\"object\": \"game controller\", \"relation\": \"side_below\", \"distance_cm\": 9.1}]}", "target": "tv remote", "split": "train"}
{"id": "9b74afd2d2_0", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "9b74afd2d2_1", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. monitor\nB. box\nC. office chair\nD. snack bag", "choices": ["A. monitor", "B. box", "C. office chair", "D. snack bag"], "answer": "C. office chair", "target": "backpack", "split": "train"}
{"id": "9b74afd2d2_2", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the office chair, about 30 cm away. It is also to the side of and below the snack bag, about 45 cm away. It is further to the side of and below the monitor, about 46 cm away.", "target": "backpack", "split": "train"}
{"id": "9b74afd2d2_3", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 30.2}, {\"object\": \"snack bag\", \"relation\": \"side_below\", \"distance_cm\": 45.0}]}", "target": "backpack", "split": "train"}
{"id": "9b74afd2d2_4", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "9b74afd2d2_5", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mug\nB. office chair\nC. heater\nD. telephone", "choices": ["A. mug", "B. office chair", "C. heater", "D. telephone"], "answer": "D. telephone", "target": "bottle", "split": "train"}
{"id": "9b74afd2d2_6", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the telephone, about 7 cm away. It is further to the side of and above the office chair, about 29 cm away.", "target": "bottle", "split": "train"}
{"id": "9b74afd2d2_7", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"telephone\", \"relation\": \"side_above\", \"distance_cm\": 7.3}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 29.3}]}", "target": "bottle", "split": "train"}
{"id": "9b74afd2d2_8", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "9b74afd2d2_9", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. monitor\nB. snack bag\nC. laptop\nD. mug", "choices": ["A. monitor", "B. snack bag", "C. laptop", "D. mug"], "answer": "A. monitor", "target": "headphones", "split": "train"}
{"id": "9b74afd2d2_10", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also under the monitor, about 3 cm away. It is further next to the mug, about 11 cm away.", "target": "headphones", "split": "train"}
{"id": "9b74afd2d2_11", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.9}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 11.5}]}", "target": "headphones", "split": "train"}
{"id": "9b74afd2d2_12", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "9b74afd2d2_13", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. backpack\nC. mouse\nD. snack bag", "choices": ["A. monitor", "B. backpack", "C. mouse", "D. snack bag"], "answer": "A. monitor", "target": "keyboard", "split": "train"}
{"id": "9b74afd2d2_14", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the monitor, about 11 cm away. It is further near the mouse, about 17 cm away.", "target": "keyboard", "split": "train"}
{"id": "9b74afd2d2_15", "scene_id": "9b74afd2d2", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.7}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 16.7}]}", "target": "keyboard", "split": "train"}
{"id": "9bfbc75700_0", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "9bfbc75700_1", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mug\nB. cabinet\nC. radiator\nD. laundry bag", "choices": ["A. mug", "B. cabinet", "C. radiator", "D. laundry bag"], "answer": "C. radiator", "target": "bottle", "split": "train"}
{"id": "9bfbc75700_2", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the window, about 6 cm away. It is also to the side of and above the radiator, about 18 cm away. It is further to the side of and above the laundry bag, about 32 cm away.", "target": "bottle", "split": "train"}
{"id": "9bfbc75700_3", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 5.8}, {\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 18.0}]}", "target": "bottle", "split": "train"}
{"id": "9bfbc75700_4", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the calculator appear?", "choices": null, "answer": "Yes", "target": "calculator", "split": "train"}
{"id": "9bfbc75700_5", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the calculator in this scene?\nA. laptop\nB. office chair\nC. mousepad\nD. paper", "choices": ["A. laptop", "B. office chair", "C. mousepad", "D. paper"], "answer": "A. laptop", "target": "calculator", "split": "train"}
{"id": "9bfbc75700_6", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the calculator in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The calculator is on the desk. It is also next to the laptop, about 7 cm away. It is further to the side of and above the office chair, about 18 cm away.", "target": "calculator", "split": "train"}
{"id": "9bfbc75700_7", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the calculator in structured JSON format.", "choices": null, "answer": "{\"target\": \"calculator\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 7.3}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 17.6}]}", "target": "calculator", "split": "train"}
{"id": "9bfbc75700_8", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "9bfbc75700_9", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. desk\nB. laptop charger\nC. office chair\nD. stool", "choices": ["A. desk", "B. laptop charger", "C. office chair", "D. stool"], "answer": "C. office chair", "target": "extension cord", "split": "train"}
{"id": "9bfbc75700_10", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the office chair, about 8 cm away. It is also near the laptop charger, about 17 cm away. It is further to the side of and below the desk, about 20 cm away.", "target": "extension cord", "split": "train"}
{"id": "9bfbc75700_11", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 7.8}, {\"object\": \"laptop charger\", \"relation\": \"near\", \"distance_cm\": 16.9}]}", "target": "extension cord", "split": "train"}
{"id": "9bfbc75700_12", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "9bfbc75700_13", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the handbag in this scene?\nA. trolley\nB. yoga mat\nC. cardboard box\nD. bag", "choices": ["A. trolley", "B. yoga mat", "C. cardboard box", "D. bag"], "answer": "A. trolley", "target": "handbag", "split": "train"}
{"id": "9bfbc75700_14", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the handbag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The handbag is to the side of and above the trolley, about 13 cm away. It is also to the side of and above the yoga mat, about 16 cm away. It is further to the side of and above the cardboard box, about 36 cm away.", "target": "handbag", "split": "train"}
{"id": "9bfbc75700_15", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the handbag in structured JSON format.", "choices": null, "answer": "{\"target\": \"handbag\", \"support_surface\": \"wall mounted table\", \"references\": [{\"object\": \"trolley\", \"relation\": \"side_above\", \"distance_cm\": 12.9}, {\"object\": \"yoga mat\", \"relation\": \"side_above\", \"distance_cm\": 15.9}]}", "target": "handbag", "split": "train"}
{"id": "9bfbc75700_16", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "9bfbc75700_17", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the mouse, about 3 cm away. It is also to the side of and above the desk, about 4 cm away. It is further to the side of and above the office chair, about 6 cm away.", "target": "headphones", "split": "train"}
{"id": "9bfbc75700_18", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 3.8}]}", "target": "headphones", "split": "train"}
{"id": "9bfbc75700_19", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "9bfbc75700_20", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. office chair\nB. laptop\nC. headphones\nD. mouse", "choices": ["A. office chair", "B. laptop", "C. headphones", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "train"}
{"id": "9bfbc75700_21", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the mouse, about 8 cm away. It is further near the headphones, about 15 cm away.", "target": "keyboard", "split": "train"}
{"id": "9bfbc75700_22", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 8.1}, {\"object\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 15.5}]}", "target": "keyboard", "split": "train"}
{"id": "9bfbc75700_23", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "9bfbc75700_24", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. radiator\nB. desk\nC. extension cord\nD. office chair", "choices": ["A. radiator", "B. desk", "C. extension cord", "D. office chair"], "answer": "B. desk", "target": "laptop charger", "split": "train"}
{"id": "9bfbc75700_25", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the desk, about 11 cm away. It is also to the side of and below the office chair, about 13 cm away. It is further near the extension cord, about 17 cm away.", "target": "laptop charger", "split": "train"}
{"id": "9bfbc75700_26", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 11.5}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 12.9}]}", "target": "laptop charger", "split": "train"}
{"id": "9bfbc75700_27", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "9bfbc75700_28", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. desk\nB. keyboard\nC. monitor\nD. headphones", "choices": ["A. desk", "B. keyboard", "C. monitor", "D. headphones"], "answer": "D. headphones", "target": "mouse", "split": "train"}
{"id": "9bfbc75700_29", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the headphones, about 3 cm away. It is also to the side of and above the desk, about 7 cm away. It is further to the side of and below the monitor, about 7 cm away.", "target": "mouse", "split": "train"}
{"id": "9bfbc75700_30", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 6.6}]}", "target": "mouse", "split": "train"}
{"id": "9bfbc75700_31", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lanyard appear?", "choices": null, "answer": "No", "target": "lanyard", "split": "train"}
{"id": "9bfbc75700_32", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. calculator\nB. office chair\nC. plush toy\nD. bookshelf", "choices": ["A. calculator", "B. office chair", "C. plush toy", "D. bookshelf"], "answer": "B. office chair", "target": "mousepad", "split": "train"}
{"id": "9bfbc75700_33", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "33", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the desk. It is also to the side of and above the office chair, about 3 cm away. It is further to the side of and below the bookshelf, about 13 cm away.", "target": "mousepad", "split": "train"}
{"id": "9bfbc75700_34", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "34", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 13.2}]}", "target": "mousepad", "split": "train"}
{"id": "9bfbc75700_35", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "35", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "9bfbc75700_36", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "36", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. belt\nB. windowsill\nC. calendar\nD. deodorant", "choices": ["A. belt", "B. windowsill", "C. calendar", "D. deodorant"], "answer": "A. belt", "target": "mug", "split": "train"}
{"id": "9bfbc75700_37", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "37", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the cabinet. It is also next to the belt, about 7 cm away. It is further next to the calendar, about 12 cm away.", "target": "mug", "split": "train"}
{"id": "9bfbc75700_38", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "38", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"belt\", \"relation\": \"next_to\", \"distance_cm\": 7.3}, {\"object\": \"calendar\", \"relation\": \"next_to\", \"distance_cm\": 11.9}]}", "target": "mug", "split": "train"}
{"id": "9bfbc75700_39", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "39", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "9bfbc75700_40", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "40", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. bag\nB. slippers\nC. fireplace\nD. pillow", "choices": ["A. bag", "B. slippers", "C. fireplace", "D. pillow"], "answer": "C. fireplace", "target": "phone charger", "split": "train"}
{"id": "9bfbc75700_41", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "41", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the fireplace, about 12 cm away. It is also to the side of and above the slippers, about 41 cm away. It is further to the side of and below the pillow, about 64 cm away.", "target": "phone charger", "split": "train"}
{"id": "9bfbc75700_42", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "42", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"fireplace\", \"relation\": \"side_below\", \"distance_cm\": 11.8}, {\"object\": \"slippers\", \"relation\": \"side_above\", \"distance_cm\": 41.0}]}", "target": "phone charger", "split": "train"}
{"id": "9bfbc75700_43", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "43", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "9bfbc75700_44", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "44", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. wardrobe\nB. bag\nC. cardboard box\nD. paper bag", "choices": ["A. wardrobe", "B. bag", "C. cardboard box", "D. paper bag"], "answer": "A. wardrobe", "target": "spray bottle", "split": "train"}
{"id": "9bfbc75700_45", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "45", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and below the wardrobe, about 14 cm away. It is also to the side of and above the paper bag, about 21 cm away. It is further to the side of and below the door, about 67 cm away.", "target": "spray bottle", "split": "train"}
{"id": "9bfbc75700_46", "scene_id": "9bfbc75700", "source_dataset": "scannetpp", "local_sample_id": "46", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"door\", \"references\": [{\"object\": \"wardrobe\", \"relation\": \"side_below\", \"distance_cm\": 14.1}, {\"object\": \"paper bag\", \"relation\": \"side_above\", \"distance_cm\": 20.5}]}", "target": "spray bottle", "split": "train"}
{"id": "9c7b4394af_0", "scene_id": "9c7b4394af", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "9c7b4394af_1", "scene_id": "9c7b4394af", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. dishwasher\nB. rice cooker\nC. utensil holder\nD. kitchen cabinet", "choices": ["A. dishwasher", "B. rice cooker", "C. utensil holder", "D. kitchen cabinet"], "answer": "B. rice cooker", "target": "power strip", "split": "train"}
{"id": "9c7b4394af_2", "scene_id": "9c7b4394af", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the rice cooker, about 7 cm away. It is also to the side of and below the utensil holder, about 13 cm away. It is further to the side of and above the kitchen cabinet, about 17 cm away.", "target": "power strip", "split": "train"}
{"id": "9c7b4394af_3", "scene_id": "9c7b4394af", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"rice cooker\", \"relation\": \"side_below\", \"distance_cm\": 6.9}, {\"object\": \"utensil holder\", \"relation\": \"side_below\", \"distance_cm\": 13.0}]}", "target": "power strip", "split": "train"}
{"id": "9cfea269dd_0", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "9cfea269dd_1", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. window frame\nB. clothes\nC. heater\nD. windowsill", "choices": ["A. window frame", "B. clothes", "C. heater", "D. windowsill"], "answer": "B. clothes", "target": "backpack", "split": "train"}
{"id": "9cfea269dd_2", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the clothes, about 10 cm away. It is also to the side of and below the window frame, about 23 cm away. It is further to the side of and above the windowsill, about 26 cm away.", "target": "backpack", "split": "train"}
{"id": "9cfea269dd_3", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"window frame\", \"references\": [{\"object\": \"clothes\", \"relation\": \"next_to\", \"distance_cm\": 10.5}, {\"object\": \"window frame\", \"relation\": \"side_below\", \"distance_cm\": 23.1}]}", "target": "backpack", "split": "train"}
{"id": "9cfea269dd_4", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "9cfea269dd_5", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. backpack\nB. poster\nC. shelf\nD. box", "choices": ["A. backpack", "B. poster", "C. shelf", "D. box"], "answer": "D. box", "target": "bottle", "split": "train"}
{"id": "9cfea269dd_6", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also next to the box, about 3 cm away. It is further to the side of and above the poster, about 77 cm away.", "target": "bottle", "split": "train"}
{"id": "9cfea269dd_7", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 2.9}, {\"object\": \"poster\", \"relation\": \"side_above\", \"distance_cm\": 76.9}]}", "target": "bottle", "split": "train"}
{"id": "9cfea269dd_8", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "9cfea269dd_9", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. standing lamp\nB. luggage\nC. monitor\nD. mouse", "choices": ["A. standing lamp", "B. luggage", "C. monitor", "D. mouse"], "answer": "C. monitor", "target": "keyboard", "split": "train"}
{"id": "9cfea269dd_10", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the monitor. It is further next to the mouse, about 5 cm away.", "target": "keyboard", "split": "train"}
{"id": "9cfea269dd_11", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.4}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.0}]}", "target": "keyboard", "split": "train"}
{"id": "9cfea269dd_12", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "9cfea269dd_13", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. luggage\nD. standing lamp", "choices": ["A. keyboard", "B. monitor", "C. luggage", "D. standing lamp"], "answer": "A. keyboard", "target": "mouse", "split": "train"}
{"id": "9cfea269dd_14", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 5 cm away. It is further to the side of and below the standing lamp, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "9cfea269dd_15", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"standing lamp\", \"relation\": \"side_below\", \"distance_cm\": 10.4}]}", "target": "mouse", "split": "train"}
{"id": "9cfea269dd_16", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "9cfea269dd_17", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. mouse\nB. bag\nC. monitor\nD. standing lamp", "choices": ["A. mouse", "B. bag", "C. monitor", "D. standing lamp"], "answer": "D. standing lamp", "target": "pen holder", "split": "train"}
{"id": "9cfea269dd_18", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also to the side of and below the standing lamp, about 10 cm away. It is further to the side of and below the monitor, about 27 cm away.", "target": "pen holder", "split": "train"}
{"id": "9cfea269dd_19", "scene_id": "9cfea269dd", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"standing lamp\", \"relation\": \"side_below\", \"distance_cm\": 10.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 26.6}]}", "target": "pen holder", "split": "train"}
{"id": "9d7e20fbec_0", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "9d7e20fbec_1", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the pouch, about 4 cm away. It is further to the side of and above the chair, about 4 cm away.", "target": "bottle", "split": "train"}
{"id": "9d7e20fbec_2", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pouch\", \"relation\": \"next_to\", \"distance_cm\": 3.8}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 4.2}]}", "target": "bottle", "split": "train"}
{"id": "9d7e20fbec_3", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "9d7e20fbec_4", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. mouse\nB. table lamp\nC. laptop\nD. office chair", "choices": ["A. mouse", "B. table lamp", "C. laptop", "D. office chair"], "answer": "C. laptop", "target": "headphones", "split": "train"}
{"id": "9d7e20fbec_5", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also next to the laptop, about 11 cm away. It is further near the mouse, about 17 cm away.", "target": "headphones", "split": "train"}
{"id": "9d7e20fbec_6", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 11.1}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 16.9}]}", "target": "headphones", "split": "train"}
{"id": "9d7e20fbec_7", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "9d7e20fbec_8", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. table lamp\nB. monitor\nC. headphones\nD. office chair", "choices": ["A. table lamp", "B. monitor", "C. headphones", "D. office chair"], "answer": "D. office chair", "target": "mouse", "split": "train"}
{"id": "9d7e20fbec_9", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and above the office chair, about 11 cm away. It is further to the side of and below the table lamp, about 16 cm away.", "target": "mouse", "split": "train"}
{"id": "9d7e20fbec_10", "scene_id": "9d7e20fbec", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 11.2}, {\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 15.9}]}", "target": "mouse", "split": "train"}
{"id": "9d8fc94966_0", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "9d8fc94966_1", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. jug\nC. power extender\nD. monitor", "choices": ["A. mouse", "B. jug", "C. power extender", "D. monitor"], "answer": "A. mouse", "target": "keyboard", "split": "train"}
{"id": "9d8fc94966_2", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 6 cm away. It is further to the side of and below the monitor, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "9d8fc94966_3", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.9}]}", "target": "keyboard", "split": "train"}
{"id": "9d8fc94966_4", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "9d8fc94966_5", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. charger head\nD. power extender", "choices": ["A. keyboard", "B. monitor", "C. charger head", "D. power extender"], "answer": "A. keyboard", "target": "mouse", "split": "train"}
{"id": "9d8fc94966_6", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 6 cm away. It is further to the side of and below the monitor, about 20 cm away.", "target": "mouse", "split": "train"}
{"id": "9d8fc94966_7", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 19.9}]}", "target": "mouse", "split": "train"}
{"id": "9d8fc94966_8", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "9d8fc94966_9", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. kettle\nB. kitchen cabinet\nC. stove\nD. plate", "choices": ["A. kettle", "B. kitchen cabinet", "C. stove", "D. plate"], "answer": "A. kettle", "target": "power strip", "split": "train"}
{"id": "9d8fc94966_10", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the kettle, about 8 cm away. It is also near the plate, about 22 cm away. It is further to the side of and above the stove, about 42 cm away.", "target": "power strip", "split": "train"}
{"id": "9d8fc94966_11", "scene_id": "9d8fc94966", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"kettle\", \"relation\": \"side_below\", \"distance_cm\": 7.8}, {\"object\": \"plate\", \"relation\": \"near\", \"distance_cm\": 22.0}]}", "target": "power strip", "split": "train"}
{"id": "9dc5ad040f_0", "scene_id": "9dc5ad040f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "No", "target": "pen", "split": "train"}
{"id": "9dc5ad040f_1", "scene_id": "9dc5ad040f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. office chair\nB. laptop\nC. monitor\nD. pouch", "choices": ["A. office chair", "B. laptop", "C. monitor", "D. pouch"], "answer": "D. pouch", "target": "mouse pad", "split": "train"}
{"id": "9dc5ad040f_2", "scene_id": "9dc5ad040f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the table. It is also next to the pouch, about 6 cm away. It is further near the laptop, about 17 cm away.", "target": "mouse pad", "split": "train"}
{"id": "9dc5ad040f_3", "scene_id": "9dc5ad040f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pouch\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 17.5}]}", "target": "mouse pad", "split": "train"}
{"id": "9ed27ac522_0", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "9ed27ac522_1", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. clothes\nB. wardrobe\nC. paper bag\nD. bag", "choices": ["A. clothes", "B. wardrobe", "C. paper bag", "D. bag"], "answer": "B. wardrobe", "target": "bottle", "split": "train"}
{"id": "9ed27ac522_2", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is above the wardrobe, about 4 cm away. It is also near the paper bag, about 16 cm away. It is further to the side of and above the bag, about 21 cm away.", "target": "bottle", "split": "train"}
{"id": "9ed27ac522_3", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"box\", \"references\": [{\"object\": \"wardrobe\", \"relation\": \"above\", \"distance_cm\": 3.5}, {\"object\": \"paper bag\", \"relation\": \"near\", \"distance_cm\": 15.8}]}", "target": "bottle", "split": "train"}
{"id": "9ed27ac522_4", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "Yes", "target": "phone stand", "split": "train"}
{"id": "9ed27ac522_5", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone stand in this scene?\nA. bottle\nB. bedside lamp\nC. wet tissue\nD. kitchen paper", "choices": ["A. bottle", "B. bedside lamp", "C. wet tissue", "D. kitchen paper"], "answer": "A. bottle", "target": "phone stand", "split": "train"}
{"id": "9ed27ac522_6", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone stand in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone stand is under the bottle. It is also under the kitchen paper, about 5 cm away. It is further to the side of and below the bedside lamp, about 12 cm away.", "target": "phone stand", "split": "train"}
{"id": "9ed27ac522_7", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone stand in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone stand\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"bottle\", \"relation\": \"under\", \"distance_cm\": 2.2}, {\"object\": \"kitchen paper\", \"relation\": \"under\", \"distance_cm\": 4.8}]}", "target": "phone stand", "split": "train"}
{"id": "9ed27ac522_8", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "9ed27ac522_9", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. bedside table\nB. phone stand\nC. wet tissue\nD. bottle", "choices": ["A. bedside table", "B. phone stand", "C. wet tissue", "D. bottle"], "answer": "A. bedside table", "target": "power strip", "split": "train"}
{"id": "9ed27ac522_10", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the bedside table, about 4 cm away. It is also to the side of and below the bottle, about 41 cm away. It is further to the side of and below the wet tissue, about 44 cm away.", "target": "power strip", "split": "train"}
{"id": "9ed27ac522_11", "scene_id": "9ed27ac522", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"bedside table\", \"relation\": \"under\", \"distance_cm\": 3.9}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 40.6}]}", "target": "power strip", "split": "train"}
{"id": "9ef704a38d_0", "scene_id": "9ef704a38d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "9ef704a38d_1", "scene_id": "9ef704a38d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. jacket\nC. laptop\nD. table", "choices": ["A. chair", "B. jacket", "C. laptop", "D. table"], "answer": "B. jacket", "target": "backpack", "split": "train"}
{"id": "9ef704a38d_2", "scene_id": "9ef704a38d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and above the jacket, about 15 cm away. It is further to the side of and above the chair, about 21 cm away.", "target": "backpack", "split": "train"}
{"id": "9ef704a38d_3", "scene_id": "9ef704a38d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 14.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 20.6}]}", "target": "backpack", "split": "train"}
{"id": "9f79564dbf_0", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "9f79564dbf_1", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. table\nC. mouse\nD. monitor", "choices": ["A. chair", "B. table", "C. mouse", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "9f79564dbf_2", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the monitor, about 4 cm away. It is further next to the mouse, about 13 cm away.", "target": "keyboard", "split": "train"}
{"id": "9f79564dbf_3", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.6}]}", "target": "keyboard", "split": "train"}
{"id": "9f79564dbf_4", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "9f79564dbf_5", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. table\nC. keyboard\nD. chair", "choices": ["A. monitor", "B. table", "C. keyboard", "D. chair"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "9f79564dbf_6", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 13 cm away. It is further to the side of and below the monitor, about 16 cm away.", "target": "mouse", "split": "train"}
{"id": "9f79564dbf_7", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 12.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 15.5}]}", "target": "mouse", "split": "train"}
{"id": "9f79564dbf_8", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "9f79564dbf_9", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. cabinet\nB. 3d printer\nC. chair\nD. paper towel", "choices": ["A. cabinet", "B. 3d printer", "C. chair", "D. paper towel"], "answer": "A. cabinet", "target": "tissue box", "split": "train"}
{"id": "9f79564dbf_10", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also to the side of and below the cabinet, about 5 cm away. It is further to the side of and below the paper towel, about 14 cm away.", "target": "tissue box", "split": "train"}
{"id": "9f79564dbf_11", "scene_id": "9f79564dbf", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 5.2}, {\"object\": \"paper towel\", \"relation\": \"side_below\", \"distance_cm\": 13.7}]}", "target": "tissue box", "split": "train"}
{"id": "a003a6585e_0", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "a003a6585e_1", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is next to the toothbrush, about 8 cm away. It is also to the side of and above the sink, about 9 cm away. It is further to the side of and below the towel, about 19 cm away.", "target": "bottle", "split": "train"}
{"id": "a003a6585e_2", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"toothbrush\", \"relation\": \"next_to\", \"distance_cm\": 8.2}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 9.1}]}", "target": "bottle", "split": "train"}
{"id": "a003a6585e_3", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "a003a6585e_4", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote control in this scene?\nA. suitcase\nB. power socket\nC. towel\nD. clothes", "choices": ["A. suitcase", "B. power socket", "C. towel", "D. clothes"], "answer": "B. power socket", "target": "remote control", "split": "train"}
{"id": "a003a6585e_5", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote control in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote control is on the table. It is also to the side of and below the power socket, about 34 cm away. It is further near the towel, about 45 cm away.", "target": "remote control", "split": "train"}
{"id": "a003a6585e_6", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote control in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote control\", \"support_surface\": \"table\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 33.6}, {\"object\": \"towel\", \"relation\": \"near\", \"distance_cm\": 44.7}]}", "target": "remote control", "split": "train"}
{"id": "a003a6585e_7", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "Yes", "target": "toothbrush", "split": "train"}
{"id": "a003a6585e_8", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. thermostat\nB. towel\nC. sink\nD. bottle", "choices": ["A. thermostat", "B. towel", "C. sink", "D. bottle"], "answer": "D. bottle", "target": "toothbrush", "split": "train"}
{"id": "a003a6585e_9", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is next to the bottle, about 8 cm away. It is also next to the towel, about 10 cm away. It is further to the side of and above the sink, about 23 cm away.", "target": "toothbrush", "split": "train"}
{"id": "a003a6585e_10", "scene_id": "a003a6585e", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 8.2}, {\"object\": \"towel\", \"relation\": \"next_to\", \"distance_cm\": 9.7}]}", "target": "toothbrush", "split": "train"}
{"id": "a05ee63164_0", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "a05ee63164_1", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. office chair\nB. monitor\nC. mouse\nD. keyboard", "choices": ["A. office chair", "B. monitor", "C. mouse", "D. keyboard"], "answer": "B. monitor", "target": "cup", "split": "train"}
{"id": "a05ee63164_2", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and below the monitor, about 10 cm away. It is further near the mouse, about 22 cm away.", "target": "cup", "split": "train"}
{"id": "a05ee63164_3", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.4}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 22.4}]}", "target": "cup", "split": "train"}
{"id": "a05ee63164_4", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "a05ee63164_5", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. computer tower\nB. bag\nC. office chair\nD. table", "choices": ["A. computer tower", "B. bag", "C. office chair", "D. table"], "answer": "B. bag", "target": "power strip", "split": "train"}
{"id": "a05ee63164_6", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the bag, about 16 cm away. It is also to the side of and below the office chair, about 21 cm away. It is further to the side of and below the table, about 32 cm away.", "target": "power strip", "split": "train"}
{"id": "a05ee63164_7", "scene_id": "a05ee63164", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 16.4}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 20.6}]}", "target": "power strip", "split": "train"}
{"id": "a08d9a2476_0", "scene_id": "a08d9a2476", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "a24f64f7fb_0", "scene_id": "a24f64f7fb", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "a24f64f7fb_1", "scene_id": "a24f64f7fb", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the speaker, about 3 cm away. It is further next to the mouse, about 4 cm away.", "target": "keyboard", "split": "train"}
{"id": "a24f64f7fb_2", "scene_id": "a24f64f7fb", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"speaker\", \"relation\": \"under\", \"distance_cm\": 3.0}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 3.9}]}", "target": "keyboard", "split": "train"}
{"id": "a24f64f7fb_3", "scene_id": "a24f64f7fb", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "a24f64f7fb_4", "scene_id": "a24f64f7fb", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. speaker\nC. keyboard\nD. monitor stand", "choices": ["A. office chair", "B. speaker", "C. keyboard", "D. monitor stand"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "a24f64f7fb_5", "scene_id": "a24f64f7fb", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 4 cm away. It is further to the side of and below the speaker, about 15 cm away.", "target": "mouse", "split": "train"}
{"id": "a24f64f7fb_6", "scene_id": "a24f64f7fb", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.9}, {\"object\": \"speaker\", \"relation\": \"side_below\", \"distance_cm\": 15.3}]}", "target": "mouse", "split": "train"}
{"id": "a29cccc784_0", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "a29cccc784_1", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. coffee table\nB. broom\nC. bag\nD. sofa", "choices": ["A. coffee table", "B. broom", "C. bag", "D. sofa"], "answer": "A. coffee table", "target": "backpack", "split": "train"}
{"id": "a29cccc784_2", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the coffee table, about 14 cm away. It is also to the side of and above the sofa, about 18 cm away. It is further to the side of and below the broom, about 25 cm away.", "target": "backpack", "split": "train"}
{"id": "a29cccc784_3", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"coffee table\", \"relation\": \"side_above\", \"distance_cm\": 13.8}, {\"object\": \"sofa\", \"relation\": \"side_above\", \"distance_cm\": 18.0}]}", "target": "backpack", "split": "train"}
{"id": "a29cccc784_4", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "a29cccc784_5", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. tablecloth\nB. stove\nC. pot\nD. lid", "choices": ["A. tablecloth", "B. stove", "C. pot", "D. lid"], "answer": "B. stove", "target": "bottle", "split": "train"}
{"id": "a29cccc784_6", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the stove, about 5 cm away. It is further to the side of and below the tablecloth, about 8 cm away.", "target": "bottle", "split": "train"}
{"id": "a29cccc784_7", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"stove\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"tablecloth\", \"relation\": \"side_below\", \"distance_cm\": 8.4}]}", "target": "bottle", "split": "train"}
{"id": "a29cccc784_8", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "Yes", "target": "headset", "split": "train"}
{"id": "a29cccc784_9", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headset in this scene?\nA. table\nB. pen holder\nC. frame\nD. book", "choices": ["A. table", "B. pen holder", "C. frame", "D. book"], "answer": "B. pen holder", "target": "headset", "split": "train"}
{"id": "a29cccc784_10", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is next to the pen holder, about 7 cm away. It is also to the side of and above the table, about 10 cm away. It is further near the frame, about 15 cm away.", "target": "headset", "split": "train"}
{"id": "a29cccc784_11", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pen holder\", \"relation\": \"next_to\", \"distance_cm\": 7.5}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 10.1}]}", "target": "headset", "split": "train"}
{"id": "a29cccc784_12", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "a29cccc784_13", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the laptop, about 5 cm away. It is further next to the book, about 6 cm away.", "target": "keyboard", "split": "train"}
{"id": "a29cccc784_14", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 5.7}]}", "target": "keyboard", "split": "train"}
{"id": "a29cccc784_15", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "a29cccc784_16", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. monitor\nC. paper\nD. keyboard", "choices": ["A. office chair", "B. monitor", "C. paper", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "a29cccc784_17", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 14 cm away. It is further to the side of and below the monitor, about 23 cm away.", "target": "mouse", "split": "train"}
{"id": "a29cccc784_18", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 13.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 22.8}]}", "target": "mouse", "split": "train"}
{"id": "a29cccc784_19", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "a29cccc784_20", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. headset\nB. paper\nC. table\nD. book", "choices": ["A. headset", "B. paper", "C. table", "D. book"], "answer": "B. paper", "target": "pen holder", "split": "train"}
{"id": "a29cccc784_21", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is next to the paper, about 5 cm away. It is also to the side of and above the table, about 6 cm away. It is further next to the headset, about 7 cm away.", "target": "pen holder", "split": "train"}
{"id": "a29cccc784_22", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 6.3}]}", "target": "pen holder", "split": "train"}
{"id": "a29cccc784_23", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "No", "target": "highlighter", "split": "train"}
{"id": "a29cccc784_24", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. sofa\nB. pillow\nC. sheet\nD. tablet", "choices": ["A. sofa", "B. pillow", "C. sheet", "D. tablet"], "answer": "A. sofa", "target": "power strip", "split": "train"}
{"id": "a29cccc784_25", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the sofa, about 16 cm away. It is also to the side of and below the sheet, about 29 cm away. It is further below the tablet, about 33 cm away.", "target": "power strip", "split": "train"}
{"id": "a29cccc784_26", "scene_id": "a29cccc784", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"sofa\", \"relation\": \"side_below\", \"distance_cm\": 15.6}, {\"object\": \"sheet\", \"relation\": \"side_below\", \"distance_cm\": 29.3}]}", "target": "power strip", "split": "train"}
{"id": "a4c043ac48_0", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "a4c043ac48_1", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. storage cabinet\nB. whiteboard eraser\nC. whiteboard\nD. adapter", "choices": ["A. storage cabinet", "B. whiteboard eraser", "C. whiteboard", "D. adapter"], "answer": "C. whiteboard", "target": "backpack", "split": "train"}
{"id": "a4c043ac48_2", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the whiteboard, about 24 cm away. It is also to the side of and below the storage cabinet, about 56 cm away. It is further to the side of and below the whiteboard eraser, about 58 cm away.", "target": "backpack", "split": "train"}
{"id": "a4c043ac48_3", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"storage cabinet\", \"references\": [{\"object\": \"whiteboard\", \"relation\": \"side_below\", \"distance_cm\": 23.9}, {\"object\": \"storage cabinet\", \"relation\": \"side_below\", \"distance_cm\": 55.6}]}", "target": "backpack", "split": "train"}
{"id": "a4c043ac48_4", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "a4c043ac48_5", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. headphones\nB. paper\nC. file cabinet\nD. trash bin", "choices": ["A. headphones", "B. paper", "C. file cabinet", "D. trash bin"], "answer": "C. file cabinet", "target": "cable", "split": "train"}
{"id": "a4c043ac48_6", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is above the file cabinet, about 6 cm away. It is also next to the paper, about 15 cm away. It is further to the side of and above the trash bin, about 18 cm away.", "target": "cable", "split": "train"}
{"id": "a4c043ac48_7", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"file cabinet\", \"relation\": \"above\", \"distance_cm\": 6.0}, {\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 14.9}]}", "target": "cable", "split": "train"}
{"id": "a4c043ac48_8", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "a4c043ac48_9", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. smartphone\nB. paper\nC. keyboard\nD. table", "choices": ["A. smartphone", "B. paper", "C. keyboard", "D. table"], "answer": "D. table", "target": "headphones", "split": "train"}
{"id": "a4c043ac48_10", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the table. It is also next to the keyboard, about 9 cm away. It is further next to the smartphone, about 10 cm away.", "target": "headphones", "split": "train"}
{"id": "a4c043ac48_11", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.0}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.9}]}", "target": "headphones", "split": "train"}
{"id": "a4c043ac48_12", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "a4c043ac48_13", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notepad in this scene?\nA. table\nB. mug\nC. file rack\nD. pens", "choices": ["A. table", "B. mug", "C. file rack", "D. pens"], "answer": "A. table", "target": "notepad", "split": "train"}
{"id": "a4c043ac48_14", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notepad is to the side of and above the table, about 3 cm away. It is also next to the mug, about 6 cm away. It is further to the side of and above the pens, about 7 cm away.", "target": "notepad", "split": "train"}
{"id": "a4c043ac48_15", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"notepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 6.1}]}", "target": "notepad", "split": "train"}
{"id": "a4c043ac48_16", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "a4c043ac48_17", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone in this scene?\nA. file cabinet\nB. books\nC. paper\nD. computer tower", "choices": ["A. file cabinet", "B. books", "C. paper", "D. computer tower"], "answer": "A. file cabinet", "target": "phone", "split": "train"}
{"id": "a4c043ac48_18", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone is to the side of and above the paper, about 10 cm away. It is also to the side of and above the computer tower, about 22 cm away. It is further to the side of and below the books, about 26 cm away.", "target": "phone", "split": "train"}
{"id": "a4c043ac48_19", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 9.8}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 22.3}]}", "target": "phone", "split": "train"}
{"id": "a4c043ac48_20", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "a4c043ac48_21", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. books\nB. file cabinet\nC. headphones\nD. paper", "choices": ["A. books", "B. file cabinet", "C. headphones", "D. paper"], "answer": "B. file cabinet", "target": "power strip", "split": "train"}
{"id": "a4c043ac48_22", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and above the file cabinet, about 6 cm away. It is also to the side of and below the books, about 20 cm away. It is further to the side of and below the headphones, about 21 cm away.", "target": "power strip", "split": "train"}
{"id": "a4c043ac48_23", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 6.1}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 19.8}]}", "target": "power strip", "split": "train"}
{"id": "a4c043ac48_24", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "Yes", "target": "smartphone", "split": "train"}
{"id": "a4c043ac48_25", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the smartphone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The smartphone is on the table. It is also next to the keyboard, about 6 cm away. It is further to the side of and below the monitor, about 9 cm away.", "target": "smartphone", "split": "train"}
{"id": "a4c043ac48_26", "scene_id": "a4c043ac48", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the smartphone in structured JSON format.", "choices": null, "answer": "{\"target\": \"smartphone\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.7}]}", "target": "smartphone", "split": "train"}
{"id": "a4d48ea6b3_0", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "a4d48ea6b3_1", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mug\nB. pillow\nC. blanket\nD. shelf", "choices": ["A. mug", "B. pillow", "C. blanket", "D. shelf"], "answer": "C. blanket", "target": "bottle", "split": "train"}
{"id": "a4d48ea6b3_2", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the nightstand. It is also to the side of and above the blanket, about 19 cm away. It is further to the side of and above the pillow, about 23 cm away.", "target": "bottle", "split": "train"}
{"id": "a4d48ea6b3_3", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"blanket\", \"relation\": \"side_above\", \"distance_cm\": 18.9}, {\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 23.2}]}", "target": "bottle", "split": "train"}
{"id": "a4d48ea6b3_4", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "a4d48ea6b3_5", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. power socket\nB. dumbbell\nC. router\nD. desk", "choices": ["A. power socket", "B. dumbbell", "C. router", "D. desk"], "answer": "B. dumbbell", "target": "cable", "split": "train"}
{"id": "a4d48ea6b3_6", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and below the dumbbell, about 6 cm away. It is also next to the router, about 11 cm away. It is further to the side of and below the desk, about 16 cm away.", "target": "cable", "split": "train"}
{"id": "a4d48ea6b3_7", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"dumbbell\", \"relation\": \"side_below\", \"distance_cm\": 5.9}, {\"object\": \"router\", \"relation\": \"next_to\", \"distance_cm\": 10.6}]}", "target": "cable", "split": "train"}
{"id": "a4d48ea6b3_8", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "a4d48ea6b3_9", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. bottle\nB. refrigerator\nC. book\nD. shelf", "choices": ["A. bottle", "B. refrigerator", "C. book", "D. shelf"], "answer": "A. bottle", "target": "mug", "split": "train"}
{"id": "a4d48ea6b3_10", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the nightstand. It is also near the bottle, about 27 cm away. It is further below the shelf, about 30 cm away.", "target": "mug", "split": "train"}
{"id": "a4d48ea6b3_11", "scene_id": "a4d48ea6b3", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"bottle\", \"relation\": \"near\", \"distance_cm\": 27.3}, {\"object\": \"shelf\", \"relation\": \"below\", \"distance_cm\": 29.8}]}", "target": "mug", "split": "train"}
{"id": "a5859cfd40_0", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "a5859cfd40_1", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. mouse\nB. room divider\nC. monitor\nD. cabinet", "choices": ["A. mouse", "B. room divider", "C. monitor", "D. cabinet"], "answer": "C. monitor", "target": "extension cord", "split": "train"}
{"id": "a5859cfd40_2", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the monitor, about 6 cm away. It is also to the side of and above the room divider, about 8 cm away. It is further to the side of and above the cabinet, about 13 cm away.", "target": "extension cord", "split": "train"}
{"id": "a5859cfd40_3", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"reception counter\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.9}, {\"object\": \"room divider\", \"relation\": \"side_above\", \"distance_cm\": 7.9}]}", "target": "extension cord", "split": "train"}
{"id": "a5859cfd40_4", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "a5859cfd40_5", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. room divider\nB. monitor\nC. cabinet\nD. mouse", "choices": ["A. room divider", "B. monitor", "C. cabinet", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "train"}
{"id": "a5859cfd40_6", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse, about 4 cm away. It is also to the side of and above the cabinet, about 12 cm away. It is further to the side of and below the monitor, about 18 cm away.", "target": "keyboard", "split": "train"}
{"id": "a5859cfd40_7", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"reception counter\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.2}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 11.9}]}", "target": "keyboard", "split": "train"}
{"id": "a5859cfd40_8", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "a5859cfd40_9", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. extension cord\nC. room divider\nD. cabinet", "choices": ["A. keyboard", "B. extension cord", "C. room divider", "D. cabinet"], "answer": "A. keyboard", "target": "mouse", "split": "train"}
{"id": "a5859cfd40_10", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 4 cm away. It is also to the side of and above the cabinet, about 15 cm away. It is further near the extension cord, about 23 cm away.", "target": "mouse", "split": "train"}
{"id": "a5859cfd40_11", "scene_id": "a5859cfd40", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"reception counter\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.2}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 14.9}]}", "target": "mouse", "split": "train"}
{"id": "a72f31d6a6_0", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "a72f31d6a6_1", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. blackboard eraser\nB. remote\nC. table\nD. telephone", "choices": ["A. blackboard eraser", "B. remote", "C. table", "D. telephone"], "answer": "C. table", "target": "bottle", "split": "train"}
{"id": "a72f31d6a6_2", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the table, about 3 cm away. It is also to the side of and above the remote, about 6 cm away. It is further to the side of and above the telephone, about 7 cm away.", "target": "bottle", "split": "train"}
{"id": "a72f31d6a6_3", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"remote\", \"relation\": \"side_above\", \"distance_cm\": 5.9}]}", "target": "bottle", "split": "train"}
{"id": "a72f31d6a6_4", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "a72f31d6a6_5", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. table\nB. bottle\nC. power socket\nD. telephone", "choices": ["A. table", "B. bottle", "C. power socket", "D. telephone"], "answer": "C. power socket", "target": "power adapter", "split": "train"}
{"id": "a72f31d6a6_6", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and above the power socket, about 12 cm away. It is also to the side of and above the table, about 18 cm away. It is further to the side of and above the telephone, about 24 cm away.", "target": "power adapter", "split": "train"}
{"id": "a72f31d6a6_7", "scene_id": "a72f31d6a6", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 11.6}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 17.7}]}", "target": "power adapter", "split": "train"}
{"id": "a8bf42d646_0", "scene_id": "a8bf42d646", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "a8bf42d646_1", "scene_id": "a8bf42d646", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power plug in this scene?\nA. machine\nB. fire extinguisher\nC. light switch\nD. picture", "choices": ["A. machine", "B. fire extinguisher", "C. light switch", "D. picture"], "answer": "D. picture", "target": "power plug", "split": "train"}
{"id": "a8bf42d646_2", "scene_id": "a8bf42d646", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power plug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power plug is to the side of and below the picture, about 25 cm away. It is also to the side of and above the machine, about 27 cm away. It is further to the side of and below the fire extinguisher, about 56 cm away.", "target": "power plug", "split": "train"}
{"id": "a8bf42d646_3", "scene_id": "a8bf42d646", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power plug in structured JSON format.", "choices": null, "answer": "{\"target\": \"power plug\", \"support_surface\": \"machine\", \"references\": [{\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 24.9}, {\"object\": \"machine\", \"relation\": \"side_above\", \"distance_cm\": 26.7}]}", "target": "power plug", "split": "train"}
{"id": "a9e4791c7e_0", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "a9e4791c7e_1", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. decoration\nB. bag\nC. baseball cap\nD. box", "choices": ["A. decoration", "B. bag", "C. baseball cap", "D. box"], "answer": "D. box", "target": "headphones", "split": "train"}
{"id": "a9e4791c7e_2", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is near the box, about 25 cm away. It is also to the side of and below the decoration, about 40 cm away. It is further to the side of and below the baseball cap, about 41 cm away.", "target": "headphones", "split": "train"}
{"id": "a9e4791c7e_3", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"box\", \"relation\": \"near\", \"distance_cm\": 25.3}, {\"object\": \"decoration\", \"relation\": \"side_below\", \"distance_cm\": 39.8}]}", "target": "headphones", "split": "train"}
{"id": "a9e4791c7e_4", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "a9e4791c7e_5", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. speaker\nB. box\nC. decoration\nD. mouse", "choices": ["A. speaker", "B. box", "C. decoration", "D. mouse"], "answer": "A. speaker", "target": "keyboard", "split": "train"}
{"id": "a9e4791c7e_6", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the speaker, about 4 cm away. It is also to the side of and below the decoration, about 6 cm away. It is further next to the mouse, about 8 cm away.", "target": "keyboard", "split": "train"}
{"id": "a9e4791c7e_7", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"speaker\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"decoration\", \"relation\": \"side_below\", \"distance_cm\": 6.0}]}", "target": "keyboard", "split": "train"}
{"id": "a9e4791c7e_8", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "a9e4791c7e_9", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. laptop\nC. speaker\nD. nintendo switch", "choices": ["A. keyboard", "B. laptop", "C. speaker", "D. nintendo switch"], "answer": "B. laptop", "target": "mouse", "split": "train"}
{"id": "a9e4791c7e_10", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the laptop, about 5 cm away. It is also next to the keyboard, about 8 cm away. It is further to the side of and below the speaker, about 11 cm away.", "target": "mouse", "split": "train"}
{"id": "a9e4791c7e_11", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 4.9}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.7}]}", "target": "mouse", "split": "train"}
{"id": "a9e4791c7e_12", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "a9e4791c7e_13", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is next to the decorations, about 4 cm away. It is also to the side of and above the kindle, about 5 cm away. It is further to the side of and above the office table, about 6 cm away.", "target": "mug", "split": "train"}
{"id": "a9e4791c7e_14", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"decorations\", \"relation\": \"next_to\", \"distance_cm\": 4.0}, {\"object\": \"kindle\", \"relation\": \"side_above\", \"distance_cm\": 4.9}]}", "target": "mug", "split": "train"}
{"id": "a9e4791c7e_15", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "a9e4791c7e_16", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. shopping bag\nB. rice cooker\nC. backpack\nD. clothes", "choices": ["A. shopping bag", "B. rice cooker", "C. backpack", "D. clothes"], "answer": "C. backpack", "target": "power strip", "split": "train"}
{"id": "a9e4791c7e_17", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the stool. It is also under the backpack, about 5 cm away. It is further to the side of and above the shopping bag, about 9 cm away.", "target": "power strip", "split": "train"}
{"id": "a9e4791c7e_18", "scene_id": "a9e4791c7e", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"backpack\", \"relation\": \"under\", \"distance_cm\": 5.0}, {\"object\": \"shopping bag\", \"relation\": \"side_above\", \"distance_cm\": 9.2}]}", "target": "power strip", "split": "train"}
{"id": "aa852f7871_0", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "aa852f7871_1", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. ottoman\nB. bag/pouch\nC. trash can\nD. clothes", "choices": ["A. ottoman", "B. bag/pouch", "C. trash can", "D. clothes"], "answer": "C. trash can", "target": "backpack", "split": "train"}
{"id": "aa852f7871_2", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the trash can, about 4 cm away. It is also to the side of and below the ottoman, about 26 cm away. It is further to the side of and below the bag/pouch, about 32 cm away.", "target": "backpack", "split": "train"}
{"id": "aa852f7871_3", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"trash can\", \"relation\": \"next_to\", \"distance_cm\": 3.5}, {\"object\": \"ottoman\", \"relation\": \"side_below\", \"distance_cm\": 25.7}]}", "target": "backpack", "split": "train"}
{"id": "aa852f7871_4", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "aa852f7871_5", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. tray\nB. curtain\nC. coffee table\nD. single couch", "choices": ["A. tray", "B. curtain", "C. coffee table", "D. single couch"], "answer": "C. coffee table", "target": "bottle", "split": "train"}
{"id": "aa852f7871_6", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is above the coffee table, about 3 cm away. It is also to the side of and above the tray, about 5 cm away. It is further to the side of and above the single couch, about 40 cm away.", "target": "bottle", "split": "train"}
{"id": "aa852f7871_7", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"coffee table\", \"relation\": \"above\", \"distance_cm\": 3.1}, {\"object\": \"tray\", \"relation\": \"side_above\", \"distance_cm\": 4.6}]}", "target": "bottle", "split": "train"}
{"id": "aa852f7871_8", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "aa852f7871_9", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. clothes\nB. laptop\nC. power socket\nD. ottoman", "choices": ["A. clothes", "B. laptop", "C. power socket", "D. ottoman"], "answer": "B. laptop", "target": "laptop charger", "split": "train"}
{"id": "aa852f7871_10", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is on the table. It is also next to the laptop, about 11 cm away. It is further to the side of and below the clothes, about 18 cm away.", "target": "laptop charger", "split": "train"}
{"id": "aa852f7871_11", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 10.8}, {\"object\": \"clothes\", \"relation\": \"side_below\", \"distance_cm\": 17.6}]}", "target": "laptop charger", "split": "train"}
{"id": "aa852f7871_12", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "aa852f7871_13", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is below the soap bottle/toiletry, about 10 cm away. It is also to the side of and below the faucet, about 11 cm away. It is further next to the hair dryer holder, about 14 cm away.", "target": "toothbrush", "split": "train"}
{"id": "aa852f7871_14", "scene_id": "aa852f7871", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"soap bottle/toiletry\", \"relation\": \"below\", \"distance_cm\": 10.4}, {\"object\": \"faucet\", \"relation\": \"side_below\", \"distance_cm\": 10.7}]}", "target": "toothbrush", "split": "train"}
{"id": "aaa11940d3_0", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "No", "target": "highlighter", "split": "train"}
{"id": "aaa11940d3_1", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. keyboard\nB. office chair\nC. monitor\nD. jacket", "choices": ["A. keyboard", "B. office chair", "C. monitor", "D. jacket"], "answer": "A. keyboard", "target": "mug", "split": "train"}
{"id": "aaa11940d3_2", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the desk. It is also next to the keyboard, about 11 cm away. It is further to the side of and below the monitor, about 17 cm away.", "target": "mug", "split": "train"}
{"id": "aaa11940d3_3", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 10.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 16.7}]}", "target": "mug", "split": "train"}
{"id": "aaa11940d3_4", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "aaa11940d3_5", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. keyboard\nB. board\nC. chair\nD. mouse", "choices": ["A. keyboard", "B. board", "C. chair", "D. mouse"], "answer": "D. mouse", "target": "notebook", "split": "train"}
{"id": "aaa11940d3_6", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is near the mouse, about 16 cm away. It is also near the keyboard, about 28 cm away.", "target": "notebook", "split": "train"}
{"id": "aaa11940d3_7", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"office desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 15.7}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 28.2}]}", "target": "notebook", "split": "train"}
{"id": "aaa11940d3_8", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "aaa11940d3_9", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. storage cabinet\nB. guitar bag\nC. office chair\nD. chair", "choices": ["A. storage cabinet", "B. guitar bag", "C. office chair", "D. chair"], "answer": "B. guitar bag", "target": "tote bag", "split": "train"}
{"id": "aaa11940d3_10", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and below the guitar bag, about 6 cm away. It is also to the side of and above the office chair, about 16 cm away. It is further to the side of and above the chair, about 34 cm away.", "target": "tote bag", "split": "train"}
{"id": "aaa11940d3_11", "scene_id": "aaa11940d3", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"guitar bag\", \"relation\": \"side_below\", \"distance_cm\": 6.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 16.3}]}", "target": "tote bag", "split": "train"}
{"id": "aaa9c251de_0", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "aaa9c251de_1", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and above the notebook, about 9 cm away. It is further next to the ball, about 9 cm away.", "target": "backpack", "split": "train"}
{"id": "aaa9c251de_2", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"notebook\", \"relation\": \"side_above\", \"distance_cm\": 8.7}, {\"object\": \"ball\", \"relation\": \"next_to\", \"distance_cm\": 9.0}]}", "target": "backpack", "split": "train"}
{"id": "aaa9c251de_3", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "aaa9c251de_4", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is next to the ball, about 6 cm away. It is also to the side of and above the cup, about 7 cm away. It is further to the side of and above the split, about 9 cm away.", "target": "bottle", "split": "train"}
{"id": "aaa9c251de_5", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"ball\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"cup\", \"relation\": \"side_above\", \"distance_cm\": 7.1}]}", "target": "bottle", "split": "train"}
{"id": "aaa9c251de_6", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "aaa9c251de_7", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. bottle\nB. backpack\nC. box\nD. ball", "choices": ["A. bottle", "B. backpack", "C. box", "D. ball"], "answer": "D. ball", "target": "cup", "split": "train"}
{"id": "aaa9c251de_8", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is under the ball, about 3 cm away. It is also to the side of and below the bottle, about 7 cm away. It is further above the box, about 8 cm away.", "target": "cup", "split": "train"}
{"id": "aaa9c251de_9", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"ball\", \"relation\": \"under\", \"distance_cm\": 3.2}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 7.1}]}", "target": "cup", "split": "train"}
{"id": "aaa9c251de_10", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "No", "target": "game controller", "split": "train"}
{"id": "aaa9c251de_11", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. notebook\nB. laptop\nC. laptop bag\nD. bag", "choices": ["A. notebook", "B. laptop", "C. laptop bag", "D. bag"], "answer": "B. laptop", "target": "mouse", "split": "train"}
{"id": "aaa9c251de_12", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the laptop, about 3 cm away. It is further to the side of and above the bag, about 18 cm away.", "target": "mouse", "split": "train"}
{"id": "aaa9c251de_13", "scene_id": "aaa9c251de", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 2.6}, {\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 18.3}]}", "target": "mouse", "split": "train"}
{"id": "ab046f8faf_0", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "train"}
{"id": "ab046f8faf_1", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. gloves\nB. box\nC. work station\nD. storage bin", "choices": ["A. gloves", "B. box", "C. work station", "D. storage bin"], "answer": "C. work station", "target": "bottle", "split": "train"}
{"id": "ab046f8faf_2", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the work station, about 6 cm away. It is also to the side of and above the gloves, about 8 cm away. It is further to the side of and above the storage bin, about 8 cm away.", "target": "bottle", "split": "train"}
{"id": "ab046f8faf_3", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"work station\", \"relation\": \"side_above\", \"distance_cm\": 6.1}, {\"object\": \"gloves\", \"relation\": \"side_above\", \"distance_cm\": 7.6}]}", "target": "bottle", "split": "train"}
{"id": "ab046f8faf_4", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "ab046f8faf_5", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the calculator in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The calculator is next to the wood stick, about 3 cm away. It is also near the books, about 18 cm away. It is further near the cloth, about 20 cm away.", "target": "calculator", "split": "train"}
{"id": "ab046f8faf_6", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the calculator in structured JSON format.", "choices": null, "answer": "{\"target\": \"calculator\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"wood stick\", \"relation\": \"next_to\", \"distance_cm\": 2.7}, {\"object\": \"books\", \"relation\": \"near\", \"distance_cm\": 17.8}]}", "target": "calculator", "split": "train"}
{"id": "ab046f8faf_7", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "ab046f8faf_8", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. box\nB. tape dispenser\nC. oragnizer\nD. folder", "choices": ["A. box", "B. tape dispenser", "C. oragnizer", "D. folder"], "answer": "D. folder", "target": "notebook", "split": "train"}
{"id": "ab046f8faf_9", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is next to the folder, about 8 cm away. It is also to the side of and above the box, about 10 cm away. It is further to the side of and above the oragnizer, about 15 cm away.", "target": "notebook", "split": "train"}
{"id": "ab046f8faf_10", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"folder\", \"relation\": \"next_to\", \"distance_cm\": 8.1}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 9.8}]}", "target": "notebook", "split": "train"}
{"id": "ab046f8faf_11", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tape dispenser appear?", "choices": null, "answer": "Yes", "target": "tape dispenser", "split": "train"}
{"id": "ab046f8faf_12", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tape dispenser in this scene?\nA. box\nB. paper\nC. cabinet\nD. storage bin", "choices": ["A. box", "B. paper", "C. cabinet", "D. storage bin"], "answer": "A. box", "target": "tape dispenser", "split": "train"}
{"id": "ab046f8faf_13", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tape dispenser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tape dispenser is to the side of and above the cabinet, about 10 cm away. It is also below the paper, about 12 cm away. It is further next to the gloves, about 14 cm away.", "target": "tape dispenser", "split": "train"}
{"id": "ab046f8faf_14", "scene_id": "ab046f8faf", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tape dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"tape dispenser\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 9.9}, {\"object\": \"paper\", \"relation\": \"below\", \"distance_cm\": 11.7}]}", "target": "tape dispenser", "split": "train"}
{"id": "ab11145646_0", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "ab11145646_1", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothes\nB. drying rack\nC. coat\nD. storage cabinet", "choices": ["A. clothes", "B. drying rack", "C. coat", "D. storage cabinet"], "answer": "A. clothes", "target": "backpack", "split": "train"}
{"id": "ab11145646_2", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the clothes, about 3 cm away. It is also to the side of and below the storage cabinet, about 13 cm away. It is further to the side of and below the coat, about 19 cm away.", "target": "backpack", "split": "train"}
{"id": "ab11145646_3", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"storage cabinet\", \"references\": [{\"object\": \"clothes\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"storage cabinet\", \"relation\": \"side_below\", \"distance_cm\": 13.0}]}", "target": "backpack", "split": "train"}
{"id": "ab11145646_4", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "ab11145646_5", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headset in this scene?\nA. toilet paper\nB. mouse\nC. laptop\nD. table lamp", "choices": ["A. toilet paper", "B. mouse", "C. laptop", "D. table lamp"], "answer": "A. toilet paper", "target": "headset", "split": "train"}
{"id": "ab11145646_6", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is on the table. It is also next to the toilet paper, about 8 cm away. It is further to the side of and below the table lamp, about 11 cm away.", "target": "headset", "split": "train"}
{"id": "ab11145646_7", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"table\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"next_to\", \"distance_cm\": 8.2}, {\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 11.0}]}", "target": "headset", "split": "train"}
{"id": "ab11145646_8", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "ab11145646_9", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and below the laptop, about 12 cm away. It is further next to the headset, about 12 cm away.", "target": "mouse", "split": "train"}
{"id": "ab11145646_10", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 11.8}, {\"object\": \"headset\", \"relation\": \"next_to\", \"distance_cm\": 11.8}]}", "target": "mouse", "split": "train"}
{"id": "ab11145646_11", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "ab11145646_12", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. bag\nB. chrismas tree\nC. drum set\nD. plant", "choices": ["A. bag", "B. chrismas tree", "C. drum set", "D. plant"], "answer": "A. bag", "target": "mug", "split": "train"}
{"id": "ab11145646_13", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the bag, about 10 cm away. It is further to the side of and above the drum set, about 25 cm away.", "target": "mug", "split": "train"}
{"id": "ab11145646_14", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 10.1}, {\"object\": \"drum set\", \"relation\": \"side_above\", \"distance_cm\": 24.6}]}", "target": "mug", "split": "train"}
{"id": "ab11145646_15", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "ab11145646_16", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. bottle\nB. basket\nC. cup\nD. coat", "choices": ["A. bottle", "B. basket", "C. cup", "D. coat"], "answer": "C. cup", "target": "power strip", "split": "train"}
{"id": "ab11145646_17", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the shelf. It is also next to the cup, about 6 cm away. It is further to the side of and above the basket, about 20 cm away.", "target": "power strip", "split": "train"}
{"id": "ab11145646_18", "scene_id": "ab11145646", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 6.4}, {\"object\": \"basket\", \"relation\": \"side_above\", \"distance_cm\": 20.1}]}", "target": "power strip", "split": "train"}
{"id": "ab4f373966_0", "scene_id": "ab4f373966", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "ab4f373966_1", "scene_id": "ab4f373966", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. shelf\nB. office chair\nC. monitor\nD. book", "choices": ["A. shelf", "B. office chair", "C. monitor", "D. book"], "answer": "C. monitor", "target": "keyboard", "split": "train"}
{"id": "ab4f373966_2", "scene_id": "ab4f373966", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the monitor, about 7 cm away. It is also to the side of and above the office chair, about 21 cm away. It is further to the side of and below the shelf, about 37 cm away.", "target": "keyboard", "split": "train"}
{"id": "ab4f373966_3", "scene_id": "ab4f373966", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 21.3}]}", "target": "keyboard", "split": "train"}
{"id": "ab61802aec_0", "scene_id": "ab61802aec", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "ab61802aec_1", "scene_id": "ab61802aec", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the table, about 17 cm away. It is also near the sticky notes, about 18 cm away. It is further to the side of and below the monitor, about 19 cm away.", "target": "bottle", "split": "train"}
{"id": "ab61802aec_2", "scene_id": "ab61802aec", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bedside cabinet\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 16.8}, {\"object\": \"sticky notes\", \"relation\": \"near\", \"distance_cm\": 17.6}]}", "target": "bottle", "split": "train"}
{"id": "ab61802aec_3", "scene_id": "ab61802aec", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "ab61802aec_4", "scene_id": "ab61802aec", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. sticky notes\nB. table\nC. toilet paper\nD. monitor", "choices": ["A. sticky notes", "B. table", "C. toilet paper", "D. monitor"], "answer": "A. sticky notes", "target": "extension cord", "split": "train"}
{"id": "ab61802aec_5", "scene_id": "ab61802aec", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is next to the sticky notes, about 7 cm away. It is also near the table, about 15 cm away. It is further to the side of and below the monitor, about 24 cm away.", "target": "extension cord", "split": "train"}
{"id": "ab61802aec_6", "scene_id": "ab61802aec", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"bedside cabinet\", \"references\": [{\"object\": \"sticky notes\", \"relation\": \"next_to\", \"distance_cm\": 7.0}, {\"object\": \"table\", \"relation\": \"near\", \"distance_cm\": 15.1}]}", "target": "extension cord", "split": "train"}
{"id": "abdb2a6169_0", "scene_id": "abdb2a6169", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "abdb2a6169_1", "scene_id": "abdb2a6169", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. bathroom mirror shelf\nB. power socket\nC. towel radiator\nD. towel", "choices": ["A. bathroom mirror shelf", "B. power socket", "C. towel radiator", "D. towel"], "answer": "B. power socket", "target": "cable", "split": "train"}
{"id": "abdb2a6169_2", "scene_id": "abdb2a6169", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is under the power socket, about 4 cm away. It is also to the side of and below the towel, about 8 cm away. It is further to the side of and below the bathroom mirror shelf, about 16 cm away.", "target": "cable", "split": "train"}
{"id": "abdb2a6169_3", "scene_id": "abdb2a6169", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"bathroom mirror shelf\", \"references\": [{\"object\": \"power socket\", \"relation\": \"under\", \"distance_cm\": 4.5}, {\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 7.9}]}", "target": "cable", "split": "train"}
{"id": "abf29d2474_0", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "abf29d2474_1", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. desk divider\nB. cabinet\nC. office table\nD. monitor", "choices": ["A. desk divider", "B. cabinet", "C. office table", "D. monitor"], "answer": "D. monitor", "target": "cup", "split": "train"}
{"id": "abf29d2474_2", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is under the monitor, about 3 cm away. It is also to the side of and below the desk divider, about 15 cm away. It is further to the side of and above the cabinet, about 39 cm away.", "target": "cup", "split": "train"}
{"id": "abf29d2474_3", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.5}, {\"object\": \"desk divider\", \"relation\": \"side_below\", \"distance_cm\": 14.7}]}", "target": "cup", "split": "train"}
{"id": "abf29d2474_4", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "abf29d2474_5", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. office table\nB. desk divider\nC. cabinet\nD. monitor", "choices": ["A. office table", "B. desk divider", "C. cabinet", "D. monitor"], "answer": "B. desk divider", "target": "headphones", "split": "train"}
{"id": "abf29d2474_6", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is under the desk divider, about 3 cm away. It is also to the side of and below the monitor, about 23 cm away. It is further to the side of and above the cabinet, about 37 cm away.", "target": "headphones", "split": "train"}
{"id": "abf29d2474_7", "scene_id": "abf29d2474", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"desk divider\", \"relation\": \"under\", \"distance_cm\": 2.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.1}]}", "target": "headphones", "split": "train"}
{"id": "ac250f0ead_0", "scene_id": "ac250f0ead", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "ac250f0ead_1", "scene_id": "ac250f0ead", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. cap\nB. sweater\nC. desk\nD. cabinet", "choices": ["A. cap", "B. sweater", "C. desk", "D. cabinet"], "answer": "C. desk", "target": "charger", "split": "train"}
{"id": "ac250f0ead_2", "scene_id": "ac250f0ead", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is to the side of and below the desk, about 30 cm away. It is also to the side of and below the cabinet, about 41 cm away. It is further to the side of and below the sweater, about 65 cm away.", "target": "charger", "split": "train"}
{"id": "ac250f0ead_3", "scene_id": "ac250f0ead", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 30.2}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 40.9}]}", "target": "charger", "split": "train"}
{"id": "ac48a9b736_0", "scene_id": "ac48a9b736", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "ac48a9b736_1", "scene_id": "ac48a9b736", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. keyboard\nB. box\nC. chair\nD. cabinet", "choices": ["A. keyboard", "B. box", "C. chair", "D. cabinet"], "answer": "C. chair", "target": "backpack", "split": "train"}
{"id": "ac48a9b736_2", "scene_id": "ac48a9b736", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 29 cm away.", "target": "backpack", "split": "train"}
{"id": "ac48a9b736_3", "scene_id": "ac48a9b736", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 29.0}]}", "target": "backpack", "split": "train"}
{"id": "ac650ac94e_0", "scene_id": "ac650ac94e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "ac650ac94e_1", "scene_id": "ac650ac94e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. bathroom mirror\nB. toilet flush button\nC. sponge\nD. power socket", "choices": ["A. bathroom mirror", "B. toilet flush button", "C. sponge", "D. power socket"], "answer": "C. sponge", "target": "bottle", "split": "train"}
{"id": "ac650ac94e_2", "scene_id": "ac650ac94e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the tray. It is also to the side of and above the sponge, about 3 cm away. It is further to the side of and above the toilet flush button, about 17 cm away.", "target": "bottle", "split": "train"}
{"id": "ac650ac94e_3", "scene_id": "ac650ac94e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"tray\", \"references\": [{\"object\": \"sponge\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"toilet flush button\", \"relation\": \"side_above\", \"distance_cm\": 17.3}]}", "target": "bottle", "split": "train"}
{"id": "ac78eb8124_0", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "ac78eb8124_1", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the marker in this scene?\nA. whiteboard eraser\nB. chair\nC. beverage bottle\nD. power sockets", "choices": ["A. whiteboard eraser", "B. chair", "C. beverage bottle", "D. power sockets"], "answer": "A. whiteboard eraser", "target": "marker", "split": "train"}
{"id": "ac78eb8124_2", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the marker in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The marker is next to the whiteboard eraser, about 10 cm away. It is also to the side of and above the power sockets, about 27 cm away.", "target": "marker", "split": "train"}
{"id": "ac78eb8124_3", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the marker in structured JSON format.", "choices": null, "answer": "{\"target\": \"marker\", \"support_surface\": \"whiteboard eraser\", \"references\": [{\"object\": \"whiteboard eraser\", \"relation\": \"next_to\", \"distance_cm\": 10.1}, {\"object\": \"power sockets\", \"relation\": \"side_above\", \"distance_cm\": 26.9}]}", "target": "marker", "split": "train"}
{"id": "ac78eb8124_4", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "ac78eb8124_5", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also under the square plastic conduit, about 5 cm away. It is further next to the window cleaning brush, about 5 cm away.", "target": "notebook", "split": "train"}
{"id": "ac78eb8124_6", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"square plastic conduit\", \"relation\": \"under\", \"distance_cm\": 4.8}, {\"object\": \"window cleaning brush\", \"relation\": \"next_to\", \"distance_cm\": 5.3}]}", "target": "notebook", "split": "train"}
{"id": "ac78eb8124_7", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "ac78eb8124_8", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the table. It is also under the telephone, about 4 cm away. It is further next to the folder, about 5 cm away.", "target": "pen", "split": "train"}
{"id": "ac78eb8124_9", "scene_id": "ac78eb8124", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"telephone\", \"relation\": \"under\", \"distance_cm\": 4.2}, {\"object\": \"folder\", \"relation\": \"next_to\", \"distance_cm\": 4.9}]}", "target": "pen", "split": "train"}
{"id": "acd69a1746_0", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "acd69a1746_1", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. cabinet\nB. telephone\nC. monitor\nD. windowsill", "choices": ["A. cabinet", "B. telephone", "C. monitor", "D. windowsill"], "answer": "A. cabinet", "target": "mug", "split": "train"}
{"id": "acd69a1746_2", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and above the cabinet, about 21 cm away. It is further near the telephone, about 23 cm away.", "target": "mug", "split": "train"}
{"id": "acd69a1746_3", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 21.2}, {\"object\": \"telephone\", \"relation\": \"near\", \"distance_cm\": 22.7}]}", "target": "mug", "split": "train"}
{"id": "acd69a1746_4", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "acd69a1746_5", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. heater\nB. power socket\nC. split\nD. cabinet", "choices": ["A. heater", "B. power socket", "C. split", "D. cabinet"], "answer": "C. split", "target": "power strip", "split": "train"}
{"id": "acd69a1746_6", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the split, about 13 cm away. It is also to the side of and below the power socket, about 70 cm away.", "target": "power strip", "split": "train"}
{"id": "acd69a1746_7", "scene_id": "acd69a1746", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"split\", \"references\": [{\"object\": \"split\", \"relation\": \"side_below\", \"distance_cm\": 12.7}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 70.3}]}", "target": "power strip", "split": "train"}
{"id": "acd95847c5_0", "scene_id": "acd95847c5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lanyard appear?", "choices": null, "answer": "No", "target": "lanyard", "split": "train"}
{"id": "acd95847c5_1", "scene_id": "acd95847c5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. pipe\nB. box\nC. clothes\nD. heater", "choices": ["A. pipe", "B. box", "C. clothes", "D. heater"], "answer": "C. clothes", "target": "backpack", "split": "train"}
{"id": "acd95847c5_2", "scene_id": "acd95847c5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the clothes. It is also above the pipe, about 38 cm away. It is further to the side of and above the box, about 41 cm away.", "target": "backpack", "split": "train"}
{"id": "acd95847c5_3", "scene_id": "acd95847c5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"box\", \"references\": [{\"object\": \"clothes\", \"relation\": \"side_above\", \"distance_cm\": 2.2}, {\"object\": \"pipe\", \"relation\": \"above\", \"distance_cm\": 38.3}]}", "target": "backpack", "split": "train"}
{"id": "ad039341f9_0", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "ad039341f9_1", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. heater\nB. books\nC. windowsill\nD. blinds", "choices": ["A. heater", "B. books", "C. windowsill", "D. blinds"], "answer": "B. books", "target": "backpack", "split": "train"}
{"id": "ad039341f9_2", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and above the books, about 3 cm away. It is further to the side of and above the windowsill, about 9 cm away.", "target": "backpack", "split": "train"}
{"id": "ad039341f9_3", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 3.3}, {\"object\": \"windowsill\", \"relation\": \"side_above\", \"distance_cm\": 8.6}]}", "target": "backpack", "split": "train"}
{"id": "ad039341f9_4", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "ad039341f9_5", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. heater\nB. blinds\nC. books\nD. table", "choices": ["A. heater", "B. blinds", "C. books", "D. table"], "answer": "C. books", "target": "pen holder", "split": "train"}
{"id": "ad039341f9_6", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is to the side of and above the books, about 11 cm away. It is also to the side of and below the blinds, about 14 cm away. It is further to the side of and below the window, about 14 cm away.", "target": "pen holder", "split": "train"}
{"id": "ad039341f9_7", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table side basket\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 10.8}, {\"object\": \"blinds\", \"relation\": \"side_below\", \"distance_cm\": 14.1}]}", "target": "pen holder", "split": "train"}
{"id": "ad039341f9_8", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "ad039341f9_9", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. cable\nB. blanket\nC. bed\nD. table", "choices": ["A. cable", "B. blanket", "C. bed", "D. table"], "answer": "A. cable", "target": "power adapter", "split": "train"}
{"id": "ad039341f9_10", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is next to the cable, about 4 cm away. It is also to the side of and below the bed, about 6 cm away. It is further to the side of and below the table, about 9 cm away.", "target": "power adapter", "split": "train"}
{"id": "ad039341f9_11", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cable\", \"relation\": \"next_to\", \"distance_cm\": 3.7}, {\"object\": \"bed\", \"relation\": \"side_below\", \"distance_cm\": 6.4}]}", "target": "power adapter", "split": "train"}
{"id": "ad039341f9_12", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "ad039341f9_13", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. table\nB. blanket\nC. cable\nD. bed", "choices": ["A. table", "B. blanket", "C. cable", "D. bed"], "answer": "C. cable", "target": "power strip", "split": "train"}
{"id": "ad039341f9_14", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is next to the cable. It is also to the side of and below the bed, about 5 cm away. It is further to the side of and below the table, about 9 cm away.", "target": "power strip", "split": "train"}
{"id": "ad039341f9_15", "scene_id": "ad039341f9", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"cable\", \"relation\": \"next_to\", \"distance_cm\": 2.3}, {\"object\": \"bed\", \"relation\": \"side_below\", \"distance_cm\": 5.1}]}", "target": "power strip", "split": "train"}
{"id": "ad2d07fd11_0", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "ad2d07fd11_1", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. table\nB. split\nC. power strip\nD. monitor", "choices": ["A. table", "B. split", "C. power strip", "D. monitor"], "answer": "B. split", "target": "mug", "split": "train"}
{"id": "ad2d07fd11_2", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the split, about 6 cm away. It is further to the side of and below the monitor, about 18 cm away.", "target": "mug", "split": "train"}
{"id": "ad2d07fd11_3", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"split\", \"relation\": \"next_to\", \"distance_cm\": 6.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 17.6}]}", "target": "mug", "split": "train"}
{"id": "ad2d07fd11_4", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "ad2d07fd11_5", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. office chair\nB. paper\nC. monitor\nD. cabinet", "choices": ["A. office chair", "B. paper", "C. monitor", "D. cabinet"], "answer": "A. office chair", "target": "notebook", "split": "train"}
{"id": "ad2d07fd11_6", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also to the side of and above the office chair, about 14 cm away. It is further to the side of and above the paper, about 19 cm away.", "target": "notebook", "split": "train"}
{"id": "ad2d07fd11_7", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 13.6}, {\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 19.1}]}", "target": "notebook", "split": "train"}
{"id": "ad2d07fd11_8", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "ad2d07fd11_9", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor\nB. stapler\nC. paper\nD. mug", "choices": ["A. monitor", "B. stapler", "C. paper", "D. mug"], "answer": "B. stapler", "target": "power strip", "split": "train"}
{"id": "ad2d07fd11_10", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also near the stapler, about 17 cm away. It is further to the side of and below the monitor, about 29 cm away.", "target": "power strip", "split": "train"}
{"id": "ad2d07fd11_11", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"stapler\", \"relation\": \"near\", \"distance_cm\": 17.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 28.6}]}", "target": "power strip", "split": "train"}
{"id": "ad2d07fd11_12", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "Yes", "target": "stapler", "split": "train"}
{"id": "ad2d07fd11_13", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. monitor\nB. mug\nC. power strip\nD. paper", "choices": ["A. monitor", "B. mug", "C. power strip", "D. paper"], "answer": "D. paper", "target": "stapler", "split": "train"}
{"id": "ad2d07fd11_14", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is on the table. It is also next to the paper, about 9 cm away. It is further near the power strip, about 17 cm away.", "target": "stapler", "split": "train"}
{"id": "ad2d07fd11_15", "scene_id": "ad2d07fd11", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 9.5}, {\"object\": \"power strip\", \"relation\": \"near\", \"distance_cm\": 17.4}]}", "target": "stapler", "split": "train"}
{"id": "ada5304e41_0", "scene_id": "ada5304e41", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "ada5304e41_1", "scene_id": "ada5304e41", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. towel\nB. kitchen cabinet\nC. microwave\nD. kitchenware", "choices": ["A. towel", "B. kitchen cabinet", "C. microwave", "D. kitchenware"], "answer": "C. microwave", "target": "bottle", "split": "train"}
{"id": "ada5304e41_2", "scene_id": "ada5304e41", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the microwave, about 5 cm away. It is also to the side of and below the kitchenware, about 27 cm away. It is further to the side of and above the kitchen cabinet, about 32 cm away.", "target": "bottle", "split": "train"}
{"id": "ada5304e41_3", "scene_id": "ada5304e41", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"microwave\", \"relation\": \"under\", \"distance_cm\": 4.9}, {\"object\": \"kitchenware\", \"relation\": \"side_below\", \"distance_cm\": 27.1}]}", "target": "bottle", "split": "train"}
{"id": "adf4ab4a53_0", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "adf4ab4a53_1", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. book\nB. laundry bag\nC. bottle\nD. shoes", "choices": ["A. book", "B. laundry bag", "C. bottle", "D. shoes"], "answer": "C. bottle", "target": "backpack", "split": "train"}
{"id": "adf4ab4a53_2", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the bottle, about 6 cm away. It is also to the side of and below the laundry bag, about 24 cm away. It is further to the side of and above the shoes, about 53 cm away.", "target": "backpack", "split": "train"}
{"id": "adf4ab4a53_3", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"laundry bag\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 5.5}, {\"object\": \"laundry bag\", \"relation\": \"side_below\", \"distance_cm\": 24.4}]}", "target": "backpack", "split": "train"}
{"id": "adf4ab4a53_4", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "adf4ab4a53_5", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. pencil case\nB. chair\nC. monitor\nD. clothes", "choices": ["A. pencil case", "B. chair", "C. monitor", "D. clothes"], "answer": "A. pencil case", "target": "pen", "split": "train"}
{"id": "adf4ab4a53_6", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the desk. It is also next to the pencil case, about 3 cm away. It is further to the side of and below the monitor, about 24 cm away.", "target": "pen", "split": "train"}
{"id": "adf4ab4a53_7", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"pencil case\", \"relation\": \"next_to\", \"distance_cm\": 2.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.6}]}", "target": "pen", "split": "train"}
{"id": "adf4ab4a53_8", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "Yes", "target": "pencil case", "split": "train"}
{"id": "adf4ab4a53_9", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil case in this scene?\nA. curtain\nB. chair\nC. monitor\nD. pen", "choices": ["A. curtain", "B. chair", "C. monitor", "D. pen"], "answer": "D. pen", "target": "pencil case", "split": "train"}
{"id": "adf4ab4a53_10", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil case is on the desk. It is also next to the pen, about 3 cm away. It is further to the side of and below the monitor, about 20 cm away.", "target": "pencil case", "split": "train"}
{"id": "adf4ab4a53_11", "scene_id": "adf4ab4a53", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"pen\", \"relation\": \"next_to\", \"distance_cm\": 2.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 20.3}]}", "target": "pencil case", "split": "train"}
{"id": "aea84db0de_0", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "aea84db0de_1", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cushion\nB. fan\nC. pillow\nD. bean bag", "choices": ["A. cushion", "B. fan", "C. pillow", "D. bean bag"], "answer": "D. bean bag", "target": "backpack", "split": "train"}
{"id": "aea84db0de_2", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the bean bag. It is also to the side of and below the cushion, about 27 cm away. It is further to the side of and below the fan, about 27 cm away.", "target": "backpack", "split": "train"}
{"id": "aea84db0de_3", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"bean bag\", \"relation\": \"next_to\", \"distance_cm\": 2.2}, {\"object\": \"cushion\", \"relation\": \"side_below\", \"distance_cm\": 26.5}]}", "target": "backpack", "split": "train"}
{"id": "aea84db0de_4", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "aea84db0de_5", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. stuffed animal\nB. bookshelf\nC. power socket\nD. curling iron", "choices": ["A. stuffed animal", "B. bookshelf", "C. power socket", "D. curling iron"], "answer": "B. bookshelf", "target": "bottle", "split": "train"}
{"id": "aea84db0de_6", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the bookshelf, about 12 cm away. It is also to the side of and below the stuffed animal, about 14 cm away. It is further to the side of and above the power socket, about 33 cm away.", "target": "bottle", "split": "train"}
{"id": "aea84db0de_7", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 12.2}, {\"object\": \"stuffed animal\", \"relation\": \"side_below\", \"distance_cm\": 13.8}]}", "target": "bottle", "split": "train"}
{"id": "aea84db0de_8", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "aea84db0de_9", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. sink\nB. tap\nC. stove\nD. kitchen pot", "choices": ["A. sink", "B. tap", "C. stove", "D. kitchen pot"], "answer": "D. kitchen pot", "target": "cup", "split": "train"}
{"id": "aea84db0de_10", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the kitchen pot, about 4 cm away. It is also to the side of and below the tap, about 10 cm away. It is further to the side of and above the sink, about 13 cm away.", "target": "cup", "split": "train"}
{"id": "aea84db0de_11", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen pot\", \"relation\": \"next_to\", \"distance_cm\": 3.7}, {\"object\": \"tap\", \"relation\": \"side_below\", \"distance_cm\": 9.8}]}", "target": "cup", "split": "train"}
{"id": "aea84db0de_12", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "aea84db0de_13", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. book\nB. monitor\nC. box\nD. rolling chair", "choices": ["A. book", "B. monitor", "C. box", "D. rolling chair"], "answer": "B. monitor", "target": "mouse", "split": "train"}
{"id": "aea84db0de_14", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and below the monitor, about 13 cm away. It is further next to the book, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "aea84db0de_15", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.9}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 14.5}]}", "target": "mouse", "split": "train"}
{"id": "aea84db0de_16", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "aea84db0de_17", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. tap\nB. kitchen cabinet\nC. sponge\nD. cutting board", "choices": ["A. tap", "B. kitchen cabinet", "C. sponge", "D. cutting board"], "answer": "D. cutting board", "target": "spray bottle", "split": "train"}
{"id": "aea84db0de_18", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is next to the cutting board, about 11 cm away. It is also to the side of and above the sponge, about 17 cm away. It is further to the side of and below the kitchen cabinet, about 24 cm away.", "target": "spray bottle", "split": "train"}
{"id": "aea84db0de_19", "scene_id": "aea84db0de", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"cutting board\", \"relation\": \"next_to\", \"distance_cm\": 11.1}, {\"object\": \"sponge\", \"relation\": \"side_above\", \"distance_cm\": 17.5}]}", "target": "spray bottle", "split": "train"}
{"id": "afd049e07c_0", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "afd049e07c_1", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse pad\nB. book\nC. office chair\nD. monitor", "choices": ["A. mouse pad", "B. book", "C. office chair", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "train"}
{"id": "afd049e07c_2", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the monitor, about 17 cm away. It is further near the book, about 23 cm away.", "target": "keyboard", "split": "train"}
{"id": "afd049e07c_3", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 17.5}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 22.9}]}", "target": "keyboard", "split": "train"}
{"id": "afd049e07c_4", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "afd049e07c_5", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. box\nB. keyboard\nC. book\nD. monitor", "choices": ["A. box", "B. keyboard", "C. book", "D. monitor"], "answer": "C. book", "target": "mouse", "split": "train"}
{"id": "afd049e07c_6", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the book, about 9 cm away. It is further to the side of and below the monitor, about 15 cm away.", "target": "mouse", "split": "train"}
{"id": "afd049e07c_7", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 8.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 14.6}]}", "target": "mouse", "split": "train"}
{"id": "afd049e07c_8", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "afd049e07c_9", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. keyboard\nB. monitor\nC. office chair\nD. box", "choices": ["A. keyboard", "B. monitor", "C. office chair", "D. box"], "answer": "B. monitor", "target": "mouse pad", "split": "train"}
{"id": "afd049e07c_10", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the table. It is also to the side of and below the monitor, about 11 cm away. It is further near the keyboard, about 26 cm away.", "target": "mouse pad", "split": "train"}
{"id": "afd049e07c_11", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.9}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 25.5}]}", "target": "mouse pad", "split": "train"}
{"id": "afd049e07c_12", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "afd049e07c_13", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. keyboard\nB. cabinet\nC. desk lamp\nD. telephone", "choices": ["A. keyboard", "B. cabinet", "C. desk lamp", "D. telephone"], "answer": "B. cabinet", "target": "power strip", "split": "train"}
{"id": "afd049e07c_14", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the cabinet, about 8 cm away. It is also to the side of and below the desk lamp, about 69 cm away. It is further to the side of and below the telephone, about 74 cm away.", "target": "power strip", "split": "train"}
{"id": "afd049e07c_15", "scene_id": "afd049e07c", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 7.6}, {\"object\": \"desk lamp\", \"relation\": \"side_below\", \"distance_cm\": 68.5}]}", "target": "power strip", "split": "train"}
{"id": "b06bcddf42_0", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "b06bcddf42_1", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bottle\nB. electronic keyboard\nC. books\nD. camera", "choices": ["A. bottle", "B. electronic keyboard", "C. books", "D. camera"], "answer": "C. books", "target": "backpack", "split": "train"}
{"id": "b06bcddf42_2", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the books, about 9 cm away. It is also to the side of and above the electronic keyboard, about 24 cm away. It is further to the side of and above the camera, about 29 cm away.", "target": "backpack", "split": "train"}
{"id": "b06bcddf42_3", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 9.3}, {\"object\": \"electronic keyboard\", \"relation\": \"side_above\", \"distance_cm\": 24.1}]}", "target": "backpack", "split": "train"}
{"id": "b06bcddf42_4", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "b06bcddf42_5", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. camera\nB. speaker\nC. books\nD. backpack", "choices": ["A. camera", "B. speaker", "C. books", "D. backpack"], "answer": "C. books", "target": "bottle", "split": "train"}
{"id": "b06bcddf42_6", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the books, about 8 cm away. It is also to the side of and above the speaker, about 25 cm away. It is further to the side of and below the backpack, about 42 cm away.", "target": "bottle", "split": "train"}
{"id": "b06bcddf42_7", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 7.8}, {\"object\": \"speaker\", \"relation\": \"side_above\", \"distance_cm\": 24.9}]}", "target": "bottle", "split": "train"}
{"id": "b06bcddf42_8", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "No", "target": "pen", "split": "train"}
{"id": "b06bcddf42_9", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. pen holder\nB. table\nC. desk lamp\nD. headphones", "choices": ["A. pen holder", "B. table", "C. desk lamp", "D. headphones"], "answer": "B. table", "target": "glasses case", "split": "train"}
{"id": "b06bcddf42_10", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is on the table. It is also next to the headphones, about 10 cm away. It is further to the side of and below the pen holder, about 13 cm away.", "target": "glasses case", "split": "train"}
{"id": "b06bcddf42_11", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 10.2}, {\"object\": \"pen holder\", \"relation\": \"side_below\", \"distance_cm\": 13.5}]}", "target": "glasses case", "split": "train"}
{"id": "b06bcddf42_12", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "b06bcddf42_13", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. table\nB. pen holder\nC. glasses case\nD. chair", "choices": ["A. table", "B. pen holder", "C. glasses case", "D. chair"], "answer": "A. table", "target": "headphones", "split": "train"}
{"id": "b06bcddf42_14", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also next to the glasses case, about 10 cm away. It is further next to the pen holder, about 15 cm away.", "target": "headphones", "split": "train"}
{"id": "b06bcddf42_15", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"glasses case\", \"relation\": \"next_to\", \"distance_cm\": 10.2}, {\"object\": \"pen holder\", \"relation\": \"next_to\", \"distance_cm\": 14.5}]}", "target": "headphones", "split": "train"}
{"id": "b06bcddf42_16", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "b06bcddf42_17", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. electric guitar\nB. audio mixer\nC. glasses case\nD. power sockets", "choices": ["A. electric guitar", "B. audio mixer", "C. glasses case", "D. power sockets"], "answer": "A. electric guitar", "target": "notebook", "split": "train"}
{"id": "b06bcddf42_18", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also to the side of and above the electric guitar, about 12 cm away. It is further next to the audio mixer, about 14 cm away.", "target": "notebook", "split": "train"}
{"id": "b06bcddf42_19", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"electric guitar\", \"relation\": \"side_above\", \"distance_cm\": 12.4}, {\"object\": \"audio mixer\", \"relation\": \"next_to\", \"distance_cm\": 14.2}]}", "target": "notebook", "split": "train"}
{"id": "b06bcddf42_20", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "b06bcddf42_21", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. headphones\nB. desk lamp\nC. chair\nD. glasses case", "choices": ["A. headphones", "B. desk lamp", "C. chair", "D. glasses case"], "answer": "B. desk lamp", "target": "pen holder", "split": "train"}
{"id": "b06bcddf42_22", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also next to the desk lamp, about 5 cm away. It is further to the side of and above the glasses case, about 13 cm away.", "target": "pen holder", "split": "train"}
{"id": "b06bcddf42_23", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"desk lamp\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"glasses case\", \"relation\": \"side_above\", \"distance_cm\": 13.5}]}", "target": "pen holder", "split": "train"}
{"id": "b06bcddf42_24", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "b06bcddf42_25", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. glasses case\nB. power sockets\nC. electric guitar\nD. chair", "choices": ["A. glasses case", "B. power sockets", "C. electric guitar", "D. chair"], "answer": "B. power sockets", "target": "power strip", "split": "train"}
{"id": "b06bcddf42_26", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is next to the rug, about 7 cm away. It is also to the side of and below the power sockets, about 35 cm away. It is further to the side of and below the chair, about 39 cm away.", "target": "power strip", "split": "train"}
{"id": "b06bcddf42_27", "scene_id": "b06bcddf42", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"rug\", \"relation\": \"next_to\", \"distance_cm\": 7.3}, {\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 34.9}]}", "target": "power strip", "split": "train"}
{"id": "b08a908f0f_0", "scene_id": "b08a908f0f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "b08a908f0f_1", "scene_id": "b08a908f0f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the bottle, about 8 cm away. It is further next to the telephone, about 8 cm away.", "target": "mug", "split": "train"}
{"id": "b08a908f0f_2", "scene_id": "b08a908f0f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 7.9}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 8.1}]}", "target": "mug", "split": "train"}
{"id": "b08a908f0f_3", "scene_id": "b08a908f0f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "b08a908f0f_4", "scene_id": "b08a908f0f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. paper\nB. mouse\nC. cabinet\nD. keyboard", "choices": ["A. paper", "B. mouse", "C. cabinet", "D. keyboard"], "answer": "B. mouse", "target": "notebook", "split": "train"}
{"id": "b08a908f0f_5", "scene_id": "b08a908f0f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also next to the mouse, about 12 cm away. It is further to the side of and above the cabinet, about 13 cm away.", "target": "notebook", "split": "train"}
{"id": "b08a908f0f_6", "scene_id": "b08a908f0f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.3}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 13.3}]}", "target": "notebook", "split": "train"}
{"id": "b09431c547_0", "scene_id": "b09431c547", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "b09431c547_1", "scene_id": "b09431c547", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. savon\nC. towel\nD. shower wall", "choices": ["A. monitor", "B. savon", "C. towel", "D. shower wall"], "answer": "B. savon", "target": "bottle", "split": "train"}
{"id": "b09431c547_2", "scene_id": "b09431c547", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the savon, about 8 cm away. It is also to the side of and below the towel, about 52 cm away.", "target": "bottle", "split": "train"}
{"id": "b09431c547_3", "scene_id": "b09431c547", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shower wall\", \"references\": [{\"object\": \"savon\", \"relation\": \"side_above\", \"distance_cm\": 7.8}, {\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 51.9}]}", "target": "bottle", "split": "train"}
{"id": "b0fe0c610f_0", "scene_id": "b0fe0c610f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "b0fe0c610f_1", "scene_id": "b0fe0c610f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. box\nB. basket\nC. rolled mattress\nD. documents", "choices": ["A. box", "B. basket", "C. rolled mattress", "D. documents"], "answer": "A. box", "target": "notebook", "split": "train"}
{"id": "b0fe0c610f_2", "scene_id": "b0fe0c610f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and above the box, about 13 cm away. It is also to the side of and above the rolled mattress, about 23 cm away. It is further to the side of and above the basket, about 59 cm away.", "target": "notebook", "split": "train"}
{"id": "b0fe0c610f_3", "scene_id": "b0fe0c610f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"storage cabinet\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 13.4}, {\"object\": \"rolled mattress\", \"relation\": \"side_above\", \"distance_cm\": 22.6}]}", "target": "notebook", "split": "train"}
{"id": "b119249da7_0", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "b119249da7_1", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. box\nB. plant\nC. cloth\nD. window sill", "choices": ["A. box", "B. plant", "C. cloth", "D. window sill"], "answer": "A. box", "target": "backpack", "split": "train"}
{"id": "b119249da7_2", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the window sill, about 31 cm away. It is also to the side of and below the plant, about 38 cm away. It is further to the side of and below the cloth, about 39 cm away.", "target": "backpack", "split": "train"}
{"id": "b119249da7_3", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"window sill\", \"references\": [{\"object\": \"window sill\", \"relation\": \"side_below\", \"distance_cm\": 30.9}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 38.4}]}", "target": "backpack", "split": "train"}
{"id": "b119249da7_4", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "b119249da7_5", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. plant\nB. cloth\nC. window frame\nD. heater", "choices": ["A. plant", "B. cloth", "C. window frame", "D. heater"], "answer": "A. plant", "target": "mug", "split": "train"}
{"id": "b119249da7_6", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and below the plant, about 8 cm away. It is also to the side of and below the window, about 10 cm away. It is further to the side of and below the window frame, about 13 cm away.", "target": "mug", "split": "train"}
{"id": "b119249da7_7", "scene_id": "b119249da7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"window sill\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 8.3}, {\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 10.2}]}", "target": "mug", "split": "train"}
{"id": "b20a261fdf_0", "scene_id": "b20a261fdf", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "b20a261fdf_1", "scene_id": "b20a261fdf", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. table\nC. monitor\nD. keyboard", "choices": ["A. chair", "B. table", "C. monitor", "D. keyboard"], "answer": "C. monitor", "target": "mouse", "split": "train"}
{"id": "b20a261fdf_2", "scene_id": "b20a261fdf", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and below the monitor, about 16 cm away. It is further near the keyboard, about 23 cm away.", "target": "mouse", "split": "train"}
{"id": "b20a261fdf_3", "scene_id": "b20a261fdf", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 16.5}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 22.9}]}", "target": "mouse", "split": "train"}
{"id": "b24697b3a1_0", "scene_id": "b24697b3a1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "b24697b3a1_1", "scene_id": "b24697b3a1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. vacuum cleaner\nB. monitor\nC. cabinet\nD. cardboard box", "choices": ["A. vacuum cleaner", "B. monitor", "C. cabinet", "D. cardboard box"], "answer": "D. cardboard box", "target": "backpack", "split": "train"}
{"id": "b24697b3a1_2", "scene_id": "b24697b3a1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is near the cardboard box, about 49 cm away.", "target": "backpack", "split": "train"}
{"id": "b24697b3a1_3", "scene_id": "b24697b3a1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cardboard box\", \"relation\": \"near\", \"distance_cm\": 48.8}]}", "target": "backpack", "split": "train"}
{"id": "b37177e6c8_0", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "b37177e6c8_1", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. power socket\nB. light switch\nC. dimmer switch\nD. cabinet", "choices": ["A. power socket", "B. light switch", "C. dimmer switch", "D. cabinet"], "answer": "B. light switch", "target": "backpack", "split": "train"}
{"id": "b37177e6c8_2", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the power socket, about 74 cm away. It is also to the side of and below the dimmer switch, about 74 cm away.", "target": "backpack", "split": "train"}
{"id": "b37177e6c8_3", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"light switch\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 74.2}, {\"object\": \"dimmer switch\", \"relation\": \"side_below\", \"distance_cm\": 74.3}]}", "target": "backpack", "split": "train"}
{"id": "b37177e6c8_4", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "b37177e6c8_5", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. cabinet\nB. dimmer switch\nC. suitcase\nD. backpack", "choices": ["A. cabinet", "B. dimmer switch", "C. suitcase", "D. backpack"], "answer": "C. suitcase", "target": "spray bottle", "split": "train"}
{"id": "b37177e6c8_6", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the cabinet. It is also to the side of and above the suitcase, about 67 cm away.", "target": "spray bottle", "split": "train"}
{"id": "b37177e6c8_7", "scene_id": "b37177e6c8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"suitcase\", \"relation\": \"side_above\", \"distance_cm\": 66.7}]}", "target": "spray bottle", "split": "train"}
{"id": "b3ac0beef0_0", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "b3ac0beef0_1", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the chair, about 17 cm away. It is also to the side of and below the coat, about 17 cm away. It is further to the side of and below the book, about 46 cm away.", "target": "backpack", "split": "train"}
{"id": "b3ac0beef0_2", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.4}, {\"object\": \"coat\", \"relation\": \"side_below\", \"distance_cm\": 17.4}]}", "target": "backpack", "split": "train"}
{"id": "b3ac0beef0_3", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "b3ac0beef0_4", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. switch\nB. towel\nC. power socket\nD. toothbrush", "choices": ["A. switch", "B. towel", "C. power socket", "D. toothbrush"], "answer": "D. toothbrush", "target": "bottle", "split": "train"}
{"id": "b3ac0beef0_5", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the toothbrush, about 34 cm away. It is also to the side of and above the towel, about 41 cm away. It is further to the side of and below the switch, about 46 cm away.", "target": "bottle", "split": "train"}
{"id": "b3ac0beef0_6", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"toothbrush\", \"relation\": \"side_below\", \"distance_cm\": 33.6}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 41.3}]}", "target": "bottle", "split": "train"}
{"id": "b3ac0beef0_7", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "b3ac0beef0_8", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office table\nB. chair\nC. window frame\nD. mug", "choices": ["A. office table", "B. chair", "C. window frame", "D. mug"], "answer": "A. office table", "target": "mouse", "split": "train"}
{"id": "b3ac0beef0_9", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the office table, about 3 cm away. It is also to the side of and below the mug, about 14 cm away. It is further to the side of and below the window frame, about 18 cm away.", "target": "mouse", "split": "train"}
{"id": "b3ac0beef0_10", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"mug\", \"relation\": \"side_below\", \"distance_cm\": 14.1}]}", "target": "mouse", "split": "train"}
{"id": "b3ac0beef0_11", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "Yes", "target": "toothbrush", "split": "train"}
{"id": "b3ac0beef0_12", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. toilet flush\nB. towel\nC. bottle\nD. sink", "choices": ["A. toilet flush", "B. towel", "C. bottle", "D. sink"], "answer": "B. towel", "target": "toothbrush", "split": "train"}
{"id": "b3ac0beef0_13", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is to the side of and above the towel, about 22 cm away. It is also to the side of and above the sink, about 26 cm away. It is further to the side of and above the bottle, about 34 cm away.", "target": "toothbrush", "split": "train"}
{"id": "b3ac0beef0_14", "scene_id": "b3ac0beef0", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"toilet paper roll\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 21.8}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 25.5}]}", "target": "toothbrush", "split": "train"}
{"id": "b4b39438f0_0", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "No", "target": "power strip", "split": "train"}
{"id": "b4b39438f0_1", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. switch\nB. bookshelf\nC. book\nD. banana", "choices": ["A. switch", "B. bookshelf", "C. book", "D. banana"], "answer": "A. switch", "target": "backpack", "split": "train"}
{"id": "b4b39438f0_2", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the switch, about 53 cm away.", "target": "backpack", "split": "train"}
{"id": "b4b39438f0_3", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"switch\", \"relation\": \"side_below\", \"distance_cm\": 52.6}]}", "target": "backpack", "split": "train"}
{"id": "b4b39438f0_4", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "b4b39438f0_5", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. headset\nB. monitor\nC. banana\nD. book", "choices": ["A. headset", "B. monitor", "C. banana", "D. book"], "answer": "A. headset", "target": "keyboard", "split": "train"}
{"id": "b4b39438f0_6", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the headset, about 3 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "keyboard", "split": "train"}
{"id": "b4b39438f0_7", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headset\", \"relation\": \"next_to\", \"distance_cm\": 3.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.0}]}", "target": "keyboard", "split": "train"}
{"id": "b4b39438f0_8", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "b4b39438f0_9", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote controller in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote controller is to the side of and above the book, about 13 cm away. It is also to the side of and above the table, about 13 cm away. It is further to the side of and above the cabinet, about 23 cm away.", "target": "remote controller", "split": "train"}
{"id": "b4b39438f0_10", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote controller in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote controller\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 12.7}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 13.2}]}", "target": "remote controller", "split": "train"}
{"id": "b4b39438f0_11", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "b4b39438f0_12", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. monitor\nB. plant pot\nC. switch\nD. headset", "choices": ["A. monitor", "B. plant pot", "C. switch", "D. headset"], "answer": "D. headset", "target": "tissue box", "split": "train"}
{"id": "b4b39438f0_13", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also near the headset, about 23 cm away. It is further to the side of and below the plant pot, about 39 cm away.", "target": "tissue box", "split": "train"}
{"id": "b4b39438f0_14", "scene_id": "b4b39438f0", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headset\", \"relation\": \"near\", \"distance_cm\": 22.6}, {\"object\": \"plant pot\", \"relation\": \"side_below\", \"distance_cm\": 38.8}]}", "target": "tissue box", "split": "train"}
{"id": "b5918e4637_0", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "b5918e4637_1", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. cabinet\nB. files\nC. headphones\nD. computer tower", "choices": ["A. cabinet", "B. files", "C. headphones", "D. computer tower"], "answer": "A. cabinet", "target": "backpack", "split": "train"}
{"id": "b5918e4637_2", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the cabinet, about 12 cm away.", "target": "backpack", "split": "train"}
{"id": "b5918e4637_3", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 12.2}]}", "target": "backpack", "split": "train"}
{"id": "b5918e4637_4", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "b5918e4637_5", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. table\nB. monitor\nC. computer tower\nD. office chair", "choices": ["A. table", "B. monitor", "C. computer tower", "D. office chair"], "answer": "A. table", "target": "headphones", "split": "train"}
{"id": "b5918e4637_6", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the table, about 3 cm away. It is also to the side of and below the monitor, about 13 cm away. It is further to the side of and above the office chair, about 15 cm away.", "target": "headphones", "split": "train"}
{"id": "b5918e4637_7", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.6}]}", "target": "headphones", "split": "train"}
{"id": "b5918e4637_8", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "b5918e4637_9", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headset in this scene?\nA. monitor\nB. mouse\nC. keyboard\nD. table", "choices": ["A. monitor", "B. mouse", "C. keyboard", "D. table"], "answer": "A. monitor", "target": "headset", "split": "train"}
{"id": "b5918e4637_10", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is on the table. It is also to the side of and below the monitor, about 5 cm away. It is further next to the mouse, about 9 cm away.", "target": "headset", "split": "train"}
{"id": "b5918e4637_11", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.2}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 9.0}]}", "target": "headset", "split": "train"}
{"id": "b5918e4637_12", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "b5918e4637_13", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. headset\nC. keyboard\nD. monitor", "choices": ["A. office chair", "B. headset", "C. keyboard", "D. monitor"], "answer": "B. headset", "target": "mouse", "split": "train"}
{"id": "b5918e4637_14", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the headset, about 9 cm away. It is further to the side of and below the monitor, about 11 cm away.", "target": "mouse", "split": "train"}
{"id": "b5918e4637_15", "scene_id": "b5918e4637", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headset\", \"relation\": \"next_to\", \"distance_cm\": 9.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 11.3}]}", "target": "mouse", "split": "train"}
{"id": "ba3f266e19_0", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "ba3f266e19_1", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. book\nB. chair\nC. table\nD. jacket", "choices": ["A. book", "B. chair", "C. table", "D. jacket"], "answer": "D. jacket", "target": "backpack", "split": "train"}
{"id": "ba3f266e19_2", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the jacket, about 37 cm away. It is also to the side of and below the table, about 40 cm away.", "target": "backpack", "split": "train"}
{"id": "ba3f266e19_3", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 37.3}, {\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 40.3}]}", "target": "backpack", "split": "train"}
{"id": "ba3f266e19_4", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "ba3f266e19_5", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. monitor\nC. book\nD. speaker", "choices": ["A. mouse", "B. monitor", "C. book", "D. speaker"], "answer": "A. mouse", "target": "keyboard", "split": "train"}
{"id": "ba3f266e19_6", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 6 cm away. It is further to the side of and below the monitor, about 12 cm away.", "target": "keyboard", "split": "train"}
{"id": "ba3f266e19_7", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 11.9}]}", "target": "keyboard", "split": "train"}
{"id": "ba3f266e19_8", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "ba3f266e19_9", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. book\nB. monitor\nC. keyboard\nD. hole puncher", "choices": ["A. book", "B. monitor", "C. keyboard", "D. hole puncher"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "ba3f266e19_10", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 6 cm away. It is further near the book, about 16 cm away.", "target": "mouse", "split": "train"}
{"id": "ba3f266e19_11", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.8}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 15.9}]}", "target": "mouse", "split": "train"}
{"id": "ba3f266e19_12", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "ba3f266e19_13", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. book\nB. laptop\nC. stapler\nD. table organizer", "choices": ["A. book", "B. laptop", "C. stapler", "D. table organizer"], "answer": "D. table organizer", "target": "pen holder", "split": "train"}
{"id": "ba3f266e19_14", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also next to the table organizer, about 8 cm away. It is further next to the book, about 12 cm away.", "target": "pen holder", "split": "train"}
{"id": "ba3f266e19_15", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table organizer\", \"relation\": \"next_to\", \"distance_cm\": 7.8}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 11.6}]}", "target": "pen holder", "split": "train"}
{"id": "ba3f266e19_16", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "ba3f266e19_17", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. speaker\nB. computer tower\nC. keyboard\nD. laptop", "choices": ["A. speaker", "B. computer tower", "C. keyboard", "D. laptop"], "answer": "A. speaker", "target": "power strip", "split": "train"}
{"id": "ba3f266e19_18", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also to the side of and below the speaker, about 23 cm away. It is further near the keyboard, about 38 cm away.", "target": "power strip", "split": "train"}
{"id": "ba3f266e19_19", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"speaker\", \"relation\": \"side_below\", \"distance_cm\": 23.2}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 38.0}]}", "target": "power strip", "split": "train"}
{"id": "ba3f266e19_20", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "ba3f266e19_21", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. paper\nB. book\nC. telephone\nD. table organizer", "choices": ["A. paper", "B. book", "C. telephone", "D. table organizer"], "answer": "A. paper", "target": "stapler", "split": "train"}
{"id": "ba3f266e19_22", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is on the table. It is also next to the paper. It is further next to the telephone, about 4 cm away.", "target": "stapler", "split": "train"}
{"id": "ba3f266e19_23", "scene_id": "ba3f266e19", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 4.5}]}", "target": "stapler", "split": "train"}
{"id": "ba8e2bce5d_0", "scene_id": "ba8e2bce5d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "ba8e2bce5d_1", "scene_id": "ba8e2bce5d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv controller in this scene?\nA. windowsill\nB. curtain\nC. heater\nD. chair", "choices": ["A. windowsill", "B. curtain", "C. heater", "D. chair"], "answer": "C. heater", "target": "tv controller", "split": "train"}
{"id": "ba8e2bce5d_2", "scene_id": "ba8e2bce5d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv controller in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv controller is to the side of and above the heater, about 11 cm away. It is also to the side of and above the chair, about 14 cm away. It is further to the side of and below the curtain, about 23 cm away.", "target": "tv controller", "split": "train"}
{"id": "ba8e2bce5d_3", "scene_id": "ba8e2bce5d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv controller in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv controller\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 11.4}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 13.9}]}", "target": "tv controller", "split": "train"}
{"id": "bac7ee3b1b_0", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "bac7ee3b1b_1", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. doorframe\nB. box\nC. trash bag\nD. light switch", "choices": ["A. doorframe", "B. box", "C. trash bag", "D. light switch"], "answer": "D. light switch", "target": "bottle", "split": "train"}
{"id": "bac7ee3b1b_2", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the light switch, about 25 cm away. It is also to the side of and above the box, about 33 cm away. It is further to the side of and above the doorframe, about 38 cm away.", "target": "bottle", "split": "train"}
{"id": "bac7ee3b1b_3", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"light switch\", \"relation\": \"side_above\", \"distance_cm\": 25.5}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 32.9}]}", "target": "bottle", "split": "train"}
{"id": "bac7ee3b1b_4", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "bac7ee3b1b_5", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. keyboard\nB. mug\nC. monitor\nD. table", "choices": ["A. keyboard", "B. mug", "C. monitor", "D. table"], "answer": "A. keyboard", "target": "cup", "split": "train"}
{"id": "bac7ee3b1b_6", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also next to the keyboard, about 3 cm away. It is further next to the mug, about 11 cm away.", "target": "cup", "split": "train"}
{"id": "bac7ee3b1b_7", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 10.7}]}", "target": "cup", "split": "train"}
{"id": "bac7ee3b1b_8", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "bac7ee3b1b_9", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. table\nB. cup\nC. mug\nD. monitor", "choices": ["A. table", "B. cup", "C. mug", "D. monitor"], "answer": "B. cup", "target": "keyboard", "split": "train"}
{"id": "bac7ee3b1b_10", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the cup, about 3 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "keyboard", "split": "train"}
{"id": "bac7ee3b1b_11", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cup\", \"relation\": \"under\", \"distance_cm\": 3.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.2}]}", "target": "keyboard", "split": "train"}
{"id": "bac7ee3b1b_12", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "No", "target": "pen", "split": "train"}
{"id": "bac7ee3b1b_13", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. cup\nB. keyboard\nC. mug\nD. laptop", "choices": ["A. cup", "B. keyboard", "C. mug", "D. laptop"], "answer": "D. laptop", "target": "mouse", "split": "train"}
{"id": "bac7ee3b1b_14", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also under the laptop, about 3 cm away. It is further near the mug, about 19 cm away.", "target": "mouse", "split": "train"}
{"id": "bac7ee3b1b_15", "scene_id": "bac7ee3b1b", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"under\", \"distance_cm\": 3.3}, {\"object\": \"mug\", \"relation\": \"near\", \"distance_cm\": 19.2}]}", "target": "mouse", "split": "train"}
{"id": "bb0ad8a081_0", "scene_id": "bb0ad8a081", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "bb0ad8a081_1", "scene_id": "bb0ad8a081", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "bb0ad8a081_2", "scene_id": "bb0ad8a081", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. light swicth\nB. door window\nC. bottle\nD. bookshelf", "choices": ["A. light swicth", "B. door window", "C. bottle", "D. bookshelf"], "answer": "B. door window", "target": "phone charger", "split": "train"}
{"id": "bb0ad8a081_3", "scene_id": "bb0ad8a081", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the door, about 9 cm away. It is also to the side of and below the door window, about 29 cm away. It is further below the light swicth, about 72 cm away.", "target": "phone charger", "split": "train"}
{"id": "bb0ad8a081_4", "scene_id": "bb0ad8a081", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 9.5}, {\"object\": \"door window\", \"relation\": \"side_below\", \"distance_cm\": 28.5}]}", "target": "phone charger", "split": "train"}
{"id": "bb87c292ad_0", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "bb87c292ad_1", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. shower arm\nB. split\nC. bath tub\nD. tap", "choices": ["A. shower arm", "B. split", "C. bath tub", "D. tap"], "answer": "A. shower arm", "target": "bottle", "split": "train"}
{"id": "bb87c292ad_2", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the shower arm, about 55 cm away. It is also to the side of and below the tap, about 69 cm away.", "target": "bottle", "split": "train"}
{"id": "bb87c292ad_3", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"split\", \"references\": [{\"object\": \"shower arm\", \"relation\": \"side_above\", \"distance_cm\": 54.5}, {\"object\": \"tap\", \"relation\": \"side_below\", \"distance_cm\": 69.1}]}", "target": "bottle", "split": "train"}
{"id": "bb87c292ad_4", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "bb87c292ad_5", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also under the monitor, about 3 cm away. It is further next to the pen holder, about 4 cm away.", "target": "cup", "split": "train"}
{"id": "bb87c292ad_6", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.4}, {\"object\": \"pen holder\", \"relation\": \"next_to\", \"distance_cm\": 4.0}]}", "target": "cup", "split": "train"}
{"id": "bb87c292ad_7", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "bb87c292ad_8", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. note\nC. headphones\nD. paper", "choices": ["A. monitor", "B. note", "C. headphones", "D. paper"], "answer": "B. note", "target": "keyboard", "split": "train"}
{"id": "bb87c292ad_9", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the note. It is further to the side of and below the monitor, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "bb87c292ad_10", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"note\", \"relation\": \"under\", \"distance_cm\": 2.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.8}]}", "target": "keyboard", "split": "train"}
{"id": "bb87c292ad_11", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "bb87c292ad_12", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. cup\nB. desk light\nC. monitor\nD. mouse", "choices": ["A. cup", "B. desk light", "C. monitor", "D. mouse"], "answer": "C. monitor", "target": "pen holder", "split": "train"}
{"id": "bb87c292ad_13", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also under the monitor, about 3 cm away. It is further next to the cup, about 4 cm away.", "target": "pen holder", "split": "train"}
{"id": "bb87c292ad_14", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.0}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 4.0}]}", "target": "pen holder", "split": "train"}
{"id": "bb87c292ad_15", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "bb87c292ad_16", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. shoe rack\nB. outlet\nC. cloth\nD. shoes", "choices": ["A. shoe rack", "B. outlet", "C. cloth", "D. shoes"], "answer": "B. outlet", "target": "power strip", "split": "train"}
{"id": "bb87c292ad_17", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the outlet, about 16 cm away. It is also to the side of and below the shoe rack, about 25 cm away. It is further to the side of and below the cloth, about 35 cm away.", "target": "power strip", "split": "train"}
{"id": "bb87c292ad_18", "scene_id": "bb87c292ad", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"shoe rack\", \"references\": [{\"object\": \"outlet\", \"relation\": \"side_below\", \"distance_cm\": 16.1}, {\"object\": \"shoe rack\", \"relation\": \"side_below\", \"distance_cm\": 24.7}]}", "target": "power strip", "split": "train"}
{"id": "bc03d88fc3_0", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "No", "target": "highlighter", "split": "train"}
{"id": "bc03d88fc3_1", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. cabinet\nB. water bottle\nC. paper\nD. table", "choices": ["A. cabinet", "B. water bottle", "C. paper", "D. table"], "answer": "C. paper", "target": "bottle", "split": "train"}
{"id": "bc03d88fc3_2", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the paper, about 3 cm away. It is further to the side of and above the cabinet, about 20 cm away.", "target": "bottle", "split": "train"}
{"id": "bc03d88fc3_3", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 3.1}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 19.9}]}", "target": "bottle", "split": "train"}
{"id": "bc03d88fc3_4", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "bc03d88fc3_5", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. paper\nB. keyboard\nC. mug\nD. water bottle", "choices": ["A. paper", "B. keyboard", "C. mug", "D. water bottle"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "bc03d88fc3_6", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard. It is further near the mug, about 15 cm away.", "target": "mouse", "split": "train"}
{"id": "bc03d88fc3_7", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"mug\", \"relation\": \"near\", \"distance_cm\": 15.1}]}", "target": "mouse", "split": "train"}
{"id": "bc03d88fc3_8", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "bc03d88fc3_9", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. cap\nB. monitor\nC. briefcase\nD. office chair", "choices": ["A. cap", "B. monitor", "C. briefcase", "D. office chair"], "answer": "C. briefcase", "target": "pen", "split": "train"}
{"id": "bc03d88fc3_10", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the table. It is also to the side of and below the briefcase, about 22 cm away. It is further to the side of and above the office chair, about 33 cm away.", "target": "pen", "split": "train"}
{"id": "bc03d88fc3_11", "scene_id": "bc03d88fc3", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"briefcase\", \"relation\": \"side_below\", \"distance_cm\": 22.3}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 33.3}]}", "target": "pen", "split": "train"}
{"id": "bc2fce1d81_0", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tape dispenser appear?", "choices": null, "answer": "No", "target": "tape dispenser", "split": "train"}
{"id": "bc2fce1d81_1", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. heater\nB. towel\nC. paper\nD. rolling trolley", "choices": ["A. heater", "B. towel", "C. paper", "D. rolling trolley"], "answer": "D. rolling trolley", "target": "power strip", "split": "train"}
{"id": "bc2fce1d81_2", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the sofa. It is also next to the rolling trolley, about 9 cm away. It is further next to the towel, about 12 cm away.", "target": "power strip", "split": "train"}
{"id": "bc2fce1d81_3", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"rolling trolley\", \"relation\": \"next_to\", \"distance_cm\": 8.8}, {\"object\": \"towel\", \"relation\": \"next_to\", \"distance_cm\": 11.7}]}", "target": "power strip", "split": "train"}
{"id": "bc2fce1d81_4", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "Yes", "target": "tv remote", "split": "train"}
{"id": "bc2fce1d81_5", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. storage box\nB. pen\nC. books\nD. towel", "choices": ["A. storage box", "B. pen", "C. books", "D. towel"], "answer": "B. pen", "target": "tv remote", "split": "train"}
{"id": "bc2fce1d81_6", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is next to the pen, about 12 cm away. It is also near the towel, about 16 cm away. It is further to the side of and above the storage box, about 17 cm away.", "target": "tv remote", "split": "train"}
{"id": "bc2fce1d81_7", "scene_id": "bc2fce1d81", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"paper\", \"references\": [{\"object\": \"pen\", \"relation\": \"next_to\", \"distance_cm\": 11.8}, {\"object\": \"towel\", \"relation\": \"near\", \"distance_cm\": 16.4}]}", "target": "tv remote", "split": "train"}
{"id": "bc94b31d5b_0", "scene_id": "bc94b31d5b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tape dispenser appear?", "choices": null, "answer": "No", "target": "tape dispenser", "split": "train"}
{"id": "bc94b31d5b_1", "scene_id": "bc94b31d5b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. chair\nB. table\nC. bag\nD. jacket", "choices": ["A. chair", "B. table", "C. bag", "D. jacket"], "answer": "C. bag", "target": "bottle", "split": "train"}
{"id": "bc94b31d5b_2", "scene_id": "bc94b31d5b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the bag, about 8 cm away. It is further to the side of and above the chair, about 22 cm away.", "target": "bottle", "split": "train"}
{"id": "bc94b31d5b_3", "scene_id": "bc94b31d5b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 8.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 21.6}]}", "target": "bottle", "split": "train"}
{"id": "be8367fcbe_0", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "be8367fcbe_1", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. baseball cap\nC. book\nD. blanket", "choices": ["A. bag", "B. baseball cap", "C. book", "D. blanket"], "answer": "A. bag", "target": "backpack", "split": "train"}
{"id": "be8367fcbe_2", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "be8367fcbe_3", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. bowl\nB. toilet paper\nC. baseball cap\nD. book", "choices": ["A. bowl", "B. toilet paper", "C. baseball cap", "D. book"], "answer": "A. bowl", "target": "cup", "split": "train"}
{"id": "be8367fcbe_4", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the shelf. It is also to the side of and above the bowl, about 19 cm away. It is further to the side of and above the toilet paper, about 29 cm away.", "target": "cup", "split": "train"}
{"id": "be8367fcbe_5", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bowl\", \"relation\": \"side_above\", \"distance_cm\": 18.9}, {\"object\": \"toilet paper\", \"relation\": \"side_above\", \"distance_cm\": 29.2}]}", "target": "cup", "split": "train"}
{"id": "be8367fcbe_6", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "be8367fcbe_7", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. power sockets\nB. notebook\nC. table\nD. pencil holder", "choices": ["A. power sockets", "B. notebook", "C. table", "D. pencil holder"], "answer": "C. table", "target": "extension cord", "split": "train"}
{"id": "be8367fcbe_8", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the table, about 30 cm away. It is also to the side of and below the power sockets, about 43 cm away. It is further to the side of and below the notebook, about 68 cm away.", "target": "extension cord", "split": "train"}
{"id": "be8367fcbe_9", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 30.1}, {\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 42.6}]}", "target": "extension cord", "split": "train"}
{"id": "be8367fcbe_10", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "be8367fcbe_11", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also under the monitor, about 5 cm away. It is further next to the mouse, about 6 cm away.", "target": "headphones", "split": "train"}
{"id": "be8367fcbe_12", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.8}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.6}]}", "target": "headphones", "split": "train"}
{"id": "be8367fcbe_13", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "be8367fcbe_14", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. power sockets\nB. chair\nC. monitor\nD. coax outlets", "choices": ["A. power sockets", "B. chair", "C. monitor", "D. coax outlets"], "answer": "D. coax outlets", "target": "laptop charger", "split": "train"}
{"id": "be8367fcbe_15", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the coax outlets, about 28 cm away. It is also to the side of and below the power sockets, about 38 cm away. It is further to the side of and below the chair, about 44 cm away.", "target": "laptop charger", "split": "train"}
{"id": "be8367fcbe_16", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"coax outlets\", \"relation\": \"side_below\", \"distance_cm\": 28.0}, {\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 38.4}]}", "target": "laptop charger", "split": "train"}
{"id": "be8367fcbe_17", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "be8367fcbe_18", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. mouse\nB. laptop\nC. chair\nD. headphones", "choices": ["A. mouse", "B. laptop", "C. chair", "D. headphones"], "answer": "D. headphones", "target": "mug", "split": "train"}
{"id": "be8367fcbe_19", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the headphones, about 7 cm away. It is further next to the mouse, about 13 cm away.", "target": "mug", "split": "train"}
{"id": "be8367fcbe_20", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 6.9}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.6}]}", "target": "mug", "split": "train"}
{"id": "be8367fcbe_21", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "Yes", "target": "pencil holder", "split": "train"}
{"id": "be8367fcbe_22", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil holder in this scene?\nA. mouse\nB. monitor\nC. notebook\nD. laptop", "choices": ["A. mouse", "B. monitor", "C. notebook", "D. laptop"], "answer": "C. notebook", "target": "pencil holder", "split": "train"}
{"id": "be8367fcbe_23", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil holder is on the table. It is also near the notebook, about 18 cm away. It is further near the mouse, about 23 cm away.", "target": "pencil holder", "split": "train"}
{"id": "be8367fcbe_24", "scene_id": "be8367fcbe", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"notebook\", \"relation\": \"near\", \"distance_cm\": 17.7}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 22.7}]}", "target": "pencil holder", "split": "train"}
{"id": "bf6e439e38_0", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "bf6e439e38_1", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. keyboard\nB. bottle\nC. monitor\nD. office chair", "choices": ["A. keyboard", "B. bottle", "C. monitor", "D. office chair"], "answer": "D. office chair", "target": "backpack", "split": "train"}
{"id": "bf6e439e38_2", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the office chair, about 22 cm away. It is also to the side of and below the monitor, about 35 cm away. It is further to the side of and below the bottle, about 38 cm away.", "target": "backpack", "split": "train"}
{"id": "bf6e439e38_3", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 21.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 34.8}]}", "target": "backpack", "split": "train"}
{"id": "bf6e439e38_4", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "bf6e439e38_5", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. keyboard\nB. monitor\nC. office chair\nD. mug", "choices": ["A. keyboard", "B. monitor", "C. office chair", "D. mug"], "answer": "D. mug", "target": "bottle", "split": "train"}
{"id": "bf6e439e38_6", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the mug, about 13 cm away. It is further to the side of and above the keyboard, about 18 cm away.", "target": "bottle", "split": "train"}
{"id": "bf6e439e38_7", "scene_id": "bf6e439e38", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 13.0}, {\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 18.3}]}", "target": "bottle", "split": "train"}
{"id": "bfcfe53c6a_0", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "bfcfe53c6a_1", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. computer cabinet\nB. dustbin\nC. electrical duct\nD. lab cabinet", "choices": ["A. computer cabinet", "B. dustbin", "C. electrical duct", "D. lab cabinet"], "answer": "B. dustbin", "target": "bottle", "split": "train"}
{"id": "bfcfe53c6a_2", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the dustbin, about 55 cm away.", "target": "bottle", "split": "train"}
{"id": "bfcfe53c6a_3", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"lab counter\", \"references\": [{\"object\": \"dustbin\", \"relation\": \"side_above\", \"distance_cm\": 54.5}]}", "target": "bottle", "split": "train"}
{"id": "bfcfe53c6a_4", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "bfcfe53c6a_5", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. electrical duct\nC. stool\nD. computer cabinet", "choices": ["A. monitor", "B. electrical duct", "C. stool", "D. computer cabinet"], "answer": "A. monitor", "target": "mouse", "split": "train"}
{"id": "bfcfe53c6a_6", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also under the monitor, about 4 cm away. It is further to the side of and below the computer cabinet, about 27 cm away.", "target": "mouse", "split": "train"}
{"id": "bfcfe53c6a_7", "scene_id": "bfcfe53c6a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.2}, {\"object\": \"computer cabinet\", \"relation\": \"side_below\", \"distance_cm\": 27.2}]}", "target": "mouse", "split": "train"}
{"id": "bfd3fd54d2_0", "scene_id": "bfd3fd54d2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "bfd3fd54d2_1", "scene_id": "bfd3fd54d2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. bag\nB. cabinet\nC. towel\nD. bed", "choices": ["A. bag", "B. cabinet", "C. towel", "D. bed"], "answer": "A. bag", "target": "cable", "split": "train"}
{"id": "bfd3fd54d2_2", "scene_id": "bfd3fd54d2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is on the cabinet. It is also to the side of and below the bag, about 6 cm away. It is further near the towel, about 27 cm away.", "target": "cable", "split": "train"}
{"id": "bfd3fd54d2_3", "scene_id": "bfd3fd54d2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"towel\", \"relation\": \"near\", \"distance_cm\": 26.5}]}", "target": "cable", "split": "train"}
{"id": "c026d108e0_0", "scene_id": "c026d108e0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "No", "target": "water bottle", "split": "train"}
{"id": "c026d108e0_1", "scene_id": "c026d108e0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. office chair\nB. chair\nC. bag\nD. jacket", "choices": ["A. office chair", "B. chair", "C. bag", "D. jacket"], "answer": "A. office chair", "target": "keyboard", "split": "train"}
{"id": "c026d108e0_2", "scene_id": "c026d108e0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the office chair, about 8 cm away. It is further to the side of and above the jacket, about 12 cm away.", "target": "keyboard", "split": "train"}
{"id": "c026d108e0_3", "scene_id": "c026d108e0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 8.3}, {\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 12.1}]}", "target": "keyboard", "split": "train"}
{"id": "c07c707449_0", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "train"}
{"id": "c07c707449_1", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. clothes\nB. clothes drying rack\nC. shoes\nD. pillow", "choices": ["A. clothes", "B. clothes drying rack", "C. shoes", "D. pillow"], "answer": "B. clothes drying rack", "target": "backpack", "split": "train"}
{"id": "c07c707449_2", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the clothes drying rack, about 21 cm away. It is also to the side of and above the shoes, about 45 cm away. It is further to the side of and below the pillow, about 50 cm away.", "target": "backpack", "split": "train"}
{"id": "c07c707449_3", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"clothes drying rack\", \"references\": [{\"object\": \"clothes drying rack\", \"relation\": \"side_below\", \"distance_cm\": 21.1}, {\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 44.7}]}", "target": "backpack", "split": "train"}
{"id": "c07c707449_4", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "c07c707449_5", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. laptop stand\nB. office chair\nC. monitor\nD. mouse", "choices": ["A. laptop stand", "B. office chair", "C. monitor", "D. mouse"], "answer": "B. office chair", "target": "headphones", "split": "train"}
{"id": "c07c707449_6", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the desk. It is also to the side of and above the office chair, about 8 cm away. It is further to the side of and below the monitor, about 20 cm away.", "target": "headphones", "split": "train"}
{"id": "c07c707449_7", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 8.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 20.4}]}", "target": "headphones", "split": "train"}
{"id": "c07c707449_8", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "c07c707449_9", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. mouse\nC. office chair\nD. papers", "choices": ["A. monitor", "B. mouse", "C. office chair", "D. papers"], "answer": "A. monitor", "target": "keyboard", "split": "train"}
{"id": "c07c707449_10", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also to the side of and below the monitor, about 6 cm away. It is further near the mouse, about 18 cm away.", "target": "keyboard", "split": "train"}
{"id": "c07c707449_11", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.2}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 18.0}]}", "target": "keyboard", "split": "train"}
{"id": "c07c707449_12", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "c07c707449_13", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. office chair\nC. papers\nD. laptop stand", "choices": ["A. keyboard", "B. office chair", "C. papers", "D. laptop stand"], "answer": "B. office chair", "target": "mouse", "split": "train"}
{"id": "c07c707449_14", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also to the side of and above the office chair, about 10 cm away. It is further next to the papers, about 13 cm away.", "target": "mouse", "split": "train"}
{"id": "c07c707449_15", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 10.0}, {\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 12.9}]}", "target": "mouse", "split": "train"}
{"id": "c07c707449_16", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "c07c707449_17", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. chair\nB. laptop stand\nC. papers\nD. table lamp", "choices": ["A. chair", "B. laptop stand", "C. papers", "D. table lamp"], "answer": "C. papers", "target": "pen holder", "split": "train"}
{"id": "c07c707449_18", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the desk. It is also to the side of and above the papers, about 5 cm away. It is further next to the laptop stand, about 12 cm away.", "target": "pen holder", "split": "train"}
{"id": "c07c707449_19", "scene_id": "c07c707449", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"papers\", \"relation\": \"side_above\", \"distance_cm\": 4.8}, {\"object\": \"laptop stand\", \"relation\": \"next_to\", \"distance_cm\": 12.4}]}", "target": "pen holder", "split": "train"}
{"id": "c08d1d52b7_0", "scene_id": "c08d1d52b7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "c08d1d52b7_1", "scene_id": "c08d1d52b7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. paper towel dispenser\nB. trash can\nC. sink\nD. shelf", "choices": ["A. paper towel dispenser", "B. trash can", "C. sink", "D. shelf"], "answer": "D. shelf", "target": "bottle", "split": "train"}
{"id": "c08d1d52b7_2", "scene_id": "c08d1d52b7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the shelf, about 17 cm away. It is also to the side of and above the trash can, about 25 cm away. It is further to the side of and below the paper towel dispenser, about 44 cm away.", "target": "bottle", "split": "train"}
{"id": "c08d1d52b7_3", "scene_id": "c08d1d52b7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 16.9}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 24.6}]}", "target": "bottle", "split": "train"}
{"id": "c0da8f4a4d_0", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "c0da8f4a4d_1", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. laptop case\nB. table\nC. candle\nD. standing lamp", "choices": ["A. laptop case", "B. table", "C. candle", "D. standing lamp"], "answer": "A. laptop case", "target": "laptop charger", "split": "train"}
{"id": "c0da8f4a4d_2", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is next to the laptop case, about 4 cm away. It is also to the side of and above the table, about 6 cm away. It is further to the side of and above the standing lamp, about 27 cm away.", "target": "laptop charger", "split": "train"}
{"id": "c0da8f4a4d_3", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop case\", \"relation\": \"next_to\", \"distance_cm\": 4.1}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 6.4}]}", "target": "laptop charger", "split": "train"}
{"id": "c0da8f4a4d_4", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "c0da8f4a4d_5", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. cabinet\nB. barstool\nC. books\nD. decoration", "choices": ["A. cabinet", "B. barstool", "C. books", "D. decoration"], "answer": "A. cabinet", "target": "mug", "split": "train"}
{"id": "c0da8f4a4d_6", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and above the cabinet, about 8 cm away. It is also next to the books, about 14 cm away. It is further to the side of and above the barstool, about 21 cm away.", "target": "mug", "split": "train"}
{"id": "c0da8f4a4d_7", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"countertop\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 8.4}, {\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 13.9}]}", "target": "mug", "split": "train"}
{"id": "c0da8f4a4d_8", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "c0da8f4a4d_9", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the smartphone in this scene?\nA. paper\nB. spectacle case\nC. cabinet\nD. decoration", "choices": ["A. paper", "B. spectacle case", "C. cabinet", "D. decoration"], "answer": "B. spectacle case", "target": "smartphone", "split": "train"}
{"id": "c0da8f4a4d_10", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the smartphone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The smartphone is next to the spectacle case, about 4 cm away. It is also to the side of and above the cabinet, about 10 cm away. It is further near the paper, about 15 cm away.", "target": "smartphone", "split": "train"}
{"id": "c0da8f4a4d_11", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the smartphone in structured JSON format.", "choices": null, "answer": "{\"target\": \"smartphone\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"spectacle case\", \"relation\": \"next_to\", \"distance_cm\": 3.6}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 9.8}]}", "target": "smartphone", "split": "train"}
{"id": "c0da8f4a4d_12", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "c0da8f4a4d_13", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. fruit hanger\nB. faucet\nC. banana\nD. sink", "choices": ["A. fruit hanger", "B. faucet", "C. banana", "D. sink"], "answer": "A. fruit hanger", "target": "tissue box", "split": "train"}
{"id": "c0da8f4a4d_14", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is next to the fruit hanger, about 6 cm away. It is also next to the banana, about 12 cm away. It is further to the side of and above the sink, about 16 cm away.", "target": "tissue box", "split": "train"}
{"id": "c0da8f4a4d_15", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"fruit hanger\", \"relation\": \"next_to\", \"distance_cm\": 6.2}, {\"object\": \"banana\", \"relation\": \"next_to\", \"distance_cm\": 11.5}]}", "target": "tissue box", "split": "train"}
{"id": "c0da8f4a4d_16", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "c0da8f4a4d_17", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. suitcase\nB. fruit hanger\nC. water bottle\nD. medicine bottle", "choices": ["A. suitcase", "B. fruit hanger", "C. water bottle", "D. medicine bottle"], "answer": "C. water bottle", "target": "tote bag", "split": "train"}
{"id": "c0da8f4a4d_18", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and above the rug, about 38 cm away. It is also to the side of and below the water bottle, about 52 cm away. It is further to the side of and below the suitcase, about 54 cm away.", "target": "tote bag", "split": "train"}
{"id": "c0da8f4a4d_19", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"rug\", \"references\": [{\"object\": \"rug\", \"relation\": \"side_above\", \"distance_cm\": 37.5}, {\"object\": \"water bottle\", \"relation\": \"side_below\", \"distance_cm\": 52.1}]}", "target": "tote bag", "split": "train"}
{"id": "c0da8f4a4d_20", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "train"}
{"id": "c0da8f4a4d_21", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. fruit hanger\nB. kitchen cabinet\nC. medicine bottle\nD. suitcase", "choices": ["A. fruit hanger", "B. kitchen cabinet", "C. medicine bottle", "D. suitcase"], "answer": "D. suitcase", "target": "water bottle", "split": "train"}
{"id": "c0da8f4a4d_22", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is to the side of and above the suitcase, about 4 cm away. It is also to the side of and above the medicine bottle, about 12 cm away. It is further to the side of and above the kitchen cabinet, about 13 cm away.", "target": "water bottle", "split": "train"}
{"id": "c0da8f4a4d_23", "scene_id": "c0da8f4a4d", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"suitcase\", \"relation\": \"side_above\", \"distance_cm\": 4.4}, {\"object\": \"medicine bottle\", \"relation\": \"side_above\", \"distance_cm\": 12.4}]}", "target": "water bottle", "split": "train"}
{"id": "c0f5742640_0", "scene_id": "c0f5742640", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "c0f5742640_1", "scene_id": "c0f5742640", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap is to the side of and below the kitchen brush, about 38 cm away. It is also to the side of and below the kitchen cabinet, about 39 cm away. It is further to the side of and above the sponge, about 42 cm away.", "target": "soap", "split": "train"}
{"id": "c0f5742640_2", "scene_id": "c0f5742640", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"kitchen brush\", \"relation\": \"side_below\", \"distance_cm\": 38.5}, {\"object\": \"kitchen cabinet\", \"relation\": \"side_below\", \"distance_cm\": 39.4}]}", "target": "soap", "split": "train"}
{"id": "c173f62b15_0", "scene_id": "c173f62b15", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "No", "target": "phone charger", "split": "train"}
{"id": "c178491bc5_0", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "c178491bc5_1", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. coax outlet\nB. beverage crate\nC. backpack\nD. bags", "choices": ["A. coax outlet", "B. beverage crate", "C. backpack", "D. bags"], "answer": "A. coax outlet", "target": "laptop charger", "split": "train"}
{"id": "c178491bc5_2", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the coax outlet, about 23 cm away.", "target": "laptop charger", "split": "train"}
{"id": "c178491bc5_3", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"power sockets\", \"references\": [{\"object\": \"coax outlet\", \"relation\": \"side_below\", \"distance_cm\": 23.5}]}", "target": "laptop charger", "split": "train"}
{"id": "c178491bc5_4", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "c178491bc5_5", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. bottle\nB. laptop\nC. chair\nD. window frame", "choices": ["A. bottle", "B. laptop", "C. chair", "D. window frame"], "answer": "B. laptop", "target": "mouse", "split": "train"}
{"id": "c178491bc5_6", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the laptop, about 7 cm away. It is further near the bottle, about 20 cm away.", "target": "mouse", "split": "train"}
{"id": "c178491bc5_7", "scene_id": "c178491bc5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"bottle\", \"relation\": \"near\", \"distance_cm\": 19.5}]}", "target": "mouse", "split": "train"}
{"id": "c24f94007b_0", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "c24f94007b_1", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. jacket\nB. cushion\nC. armchair\nD. book", "choices": ["A. jacket", "B. cushion", "C. armchair", "D. book"], "answer": "C. armchair", "target": "backpack", "split": "train"}
{"id": "c24f94007b_2", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the armchair, about 7 cm away. It is also to the side of and above the cushion, about 28 cm away. It is further to the side of and below the jacket, about 48 cm away.", "target": "backpack", "split": "train"}
{"id": "c24f94007b_3", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"armchair\", \"references\": [{\"object\": \"armchair\", \"relation\": \"next_to\", \"distance_cm\": 7.0}, {\"object\": \"cushion\", \"relation\": \"side_above\", \"distance_cm\": 27.6}]}", "target": "backpack", "split": "train"}
{"id": "c24f94007b_4", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tape dispenser appear?", "choices": null, "answer": "No", "target": "tape dispenser", "split": "train"}
{"id": "c24f94007b_5", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. speaker\nB. paper\nC. book\nD. monitor", "choices": ["A. speaker", "B. paper", "C. book", "D. monitor"], "answer": "C. book", "target": "cup", "split": "train"}
{"id": "c24f94007b_6", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also near the book, about 17 cm away. It is further near the paper, about 26 cm away.", "target": "cup", "split": "train"}
{"id": "c24f94007b_7", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 17.4}, {\"object\": \"paper\", \"relation\": \"near\", \"distance_cm\": 25.5}]}", "target": "cup", "split": "train"}
{"id": "c24f94007b_8", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "c24f94007b_9", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. mouse\nC. speaker\nD. paper", "choices": ["A. monitor", "B. mouse", "C. speaker", "D. paper"], "answer": "B. mouse", "target": "keyboard", "split": "train"}
{"id": "c24f94007b_10", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 5 cm away. It is further next to the paper, about 14 cm away.", "target": "keyboard", "split": "train"}
{"id": "c24f94007b_11", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 13.5}]}", "target": "keyboard", "split": "train"}
{"id": "c24f94007b_12", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "c24f94007b_13", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. paper\nB. mug\nC. speaker\nD. keyboard", "choices": ["A. paper", "B. mug", "C. speaker", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "c24f94007b_14", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 5 cm away. It is further next to the paper, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "c24f94007b_15", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 8.7}]}", "target": "mouse", "split": "train"}
{"id": "c24f94007b_16", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "c24f94007b_17", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. monitor\nB. mouse\nC. paper\nD. keyboard", "choices": ["A. monitor", "B. mouse", "C. paper", "D. keyboard"], "answer": "C. paper", "target": "mug", "split": "train"}
{"id": "c24f94007b_18", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and above the paper, about 4 cm away. It is further next to the mouse, about 10 cm away.", "target": "mug", "split": "train"}
{"id": "c24f94007b_19", "scene_id": "c24f94007b", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 4.4}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 10.5}]}", "target": "mug", "split": "train"}
{"id": "c2522abc00_0", "scene_id": "c2522abc00", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "No", "target": "soap", "split": "train"}
{"id": "c2522abc00_1", "scene_id": "c2522abc00", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. chair\nC. jacket\nD. table", "choices": ["A. monitor", "B. chair", "C. jacket", "D. table"], "answer": "B. chair", "target": "bottle", "split": "train"}
{"id": "c2522abc00_2", "scene_id": "c2522abc00", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the chair, about 26 cm away. It is further to the side of and above the jacket, about 29 cm away.", "target": "bottle", "split": "train"}
{"id": "c2522abc00_3", "scene_id": "c2522abc00", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 26.5}, {\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 28.8}]}", "target": "bottle", "split": "train"}
{"id": "c31ebd4b22_0", "scene_id": "c31ebd4b22", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "c31ebd4b22_1", "scene_id": "c31ebd4b22", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and above the books, about 53 cm away. It is also to the side of and below the monitor, about 54 cm away. It is further to the side of and above the computer tower, about 57 cm away.", "target": "spray bottle", "split": "train"}
{"id": "c31ebd4b22_2", "scene_id": "c31ebd4b22", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"books\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 52.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 53.7}]}", "target": "spray bottle", "split": "train"}
{"id": "c31ebd4b22_3", "scene_id": "c31ebd4b22", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "c31ebd4b22_4", "scene_id": "c31ebd4b22", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. monitor\nB. telephone\nC. paper\nD. table", "choices": ["A. monitor", "B. telephone", "C. paper", "D. table"], "answer": "D. table", "target": "tissue box", "split": "train"}
{"id": "c31ebd4b22_5", "scene_id": "c31ebd4b22", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and above the table, about 3 cm away. It is also to the side of and above the telephone, about 4 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "tissue box", "split": "train"}
{"id": "c31ebd4b22_6", "scene_id": "c31ebd4b22", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table organizer\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.9}, {\"object\": \"telephone\", \"relation\": \"side_above\", \"distance_cm\": 4.5}]}", "target": "tissue box", "split": "train"}
{"id": "c340418532_0", "scene_id": "c340418532", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "c340418532_1", "scene_id": "c340418532", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. window frame\nB. box\nC. kettle\nD. heater", "choices": ["A. window frame", "B. box", "C. kettle", "D. heater"], "answer": "B. box", "target": "power strip", "split": "train"}
{"id": "c340418532_2", "scene_id": "c340418532", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the window, about 5 cm away. It is also next to the box, about 10 cm away. It is further to the side of and below the window frame, about 30 cm away.", "target": "power strip", "split": "train"}
{"id": "c340418532_3", "scene_id": "c340418532", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"window sill\", \"references\": [{\"object\": \"window\", \"relation\": \"under\", \"distance_cm\": 5.0}, {\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 9.7}]}", "target": "power strip", "split": "train"}
{"id": "c3bc5e82c5_0", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "c3bc5e82c5_1", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the cabinet. It is also to the side of and above the bed, about 12 cm away. It is further to the side of and above the blanket, about 12 cm away.", "target": "extension cord", "split": "train"}
{"id": "c3bc5e82c5_2", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"bed\", \"relation\": \"side_above\", \"distance_cm\": 11.9}, {\"object\": \"blanket\", \"relation\": \"side_above\", \"distance_cm\": 11.9}]}", "target": "extension cord", "split": "train"}
{"id": "c3bc5e82c5_3", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "c3bc5e82c5_4", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. laptop stand\nB. bag\nC. monitor\nD. table", "choices": ["A. laptop stand", "B. bag", "C. monitor", "D. table"], "answer": "D. table", "target": "mousepad", "split": "train"}
{"id": "c3bc5e82c5_5", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is to the side of and above the table, about 5 cm away. It is also next to the bag, about 9 cm away. It is further next to the laptop stand, about 10 cm away.", "target": "mousepad", "split": "train"}
{"id": "c3bc5e82c5_6", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk mat\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.6}, {\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 9.2}]}", "target": "mousepad", "split": "train"}
{"id": "c3bc5e82c5_7", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "c3bc5e82c5_8", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. desk mat\nB. box\nC. monitor\nD. desk lamp", "choices": ["A. desk mat", "B. box", "C. monitor", "D. desk lamp"], "answer": "B. box", "target": "mug", "split": "train"}
{"id": "c3bc5e82c5_9", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the box, about 3 cm away. It is further to the side of and above the desk mat, about 13 cm away.", "target": "mug", "split": "train"}
{"id": "c3bc5e82c5_10", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"desk mat\", \"relation\": \"side_above\", \"distance_cm\": 13.5}]}", "target": "mug", "split": "train"}
{"id": "c3bc5e82c5_11", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "c3bc5e82c5_12", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. box\nB. portable heater\nC. organizer\nD. decoration", "choices": ["A. box", "B. portable heater", "C. organizer", "D. decoration"], "answer": "C. organizer", "target": "notebook", "split": "train"}
{"id": "c3bc5e82c5_13", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and below the organizer, about 17 cm away. It is also below the decoration, about 20 cm away. It is further to the side of and below the portable heater, about 24 cm away.", "target": "notebook", "split": "train"}
{"id": "c3bc5e82c5_14", "scene_id": "c3bc5e82c5", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"storage shelf\", \"references\": [{\"object\": \"organizer\", \"relation\": \"side_below\", \"distance_cm\": 16.7}, {\"object\": \"decoration\", \"relation\": \"below\", \"distance_cm\": 19.6}]}", "target": "notebook", "split": "train"}
{"id": "c40466a844_0", "scene_id": "c40466a844", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "c40466a844_1", "scene_id": "c40466a844", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. headphones\nB. laptop stand\nC. mouse\nD. monitor", "choices": ["A. headphones", "B. laptop stand", "C. mouse", "D. monitor"], "answer": "B. laptop stand", "target": "cup", "split": "train"}
{"id": "c40466a844_2", "scene_id": "c40466a844", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the laptop stand, about 5 cm away. It is also next to the mouse, about 6 cm away. It is further near the headphones, about 17 cm away.", "target": "cup", "split": "train"}
{"id": "c40466a844_3", "scene_id": "c40466a844", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"laptop stand\", \"references\": [{\"object\": \"laptop stand\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.8}]}", "target": "cup", "split": "train"}
{"id": "c47168fab2_0", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "c47168fab2_1", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. electric toothbrush\nB. shower wall\nC. sink\nD. toilet flush button", "choices": ["A. electric toothbrush", "B. shower wall", "C. sink", "D. toilet flush button"], "answer": "C. sink", "target": "bottle", "split": "train"}
{"id": "c47168fab2_2", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the sink, about 4 cm away. It is also to the side of and below the electric toothbrush, about 50 cm away. It is further to the side of and above the toilet flush button, about 60 cm away.", "target": "bottle", "split": "train"}
{"id": "c47168fab2_3", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bath counter\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 3.6}, {\"object\": \"electric toothbrush\", \"relation\": \"side_below\", \"distance_cm\": 50.4}]}", "target": "bottle", "split": "train"}
{"id": "c47168fab2_4", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "Yes", "target": "glasses case", "split": "train"}
{"id": "c47168fab2_5", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. tablet\nB. pocket calculator\nC. curtain\nD. banana", "choices": ["A. tablet", "B. pocket calculator", "C. curtain", "D. banana"], "answer": "B. pocket calculator", "target": "glasses case", "split": "train"}
{"id": "c47168fab2_6", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is to the side of and above the pocket calculator, about 9 cm away. It is also next to the tablet, about 14 cm away. It is further to the side of and below the curtain, about 23 cm away.", "target": "glasses case", "split": "train"}
{"id": "c47168fab2_7", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"window\", \"references\": [{\"object\": \"pocket calculator\", \"relation\": \"side_above\", \"distance_cm\": 9.3}, {\"object\": \"tablet\", \"relation\": \"next_to\", \"distance_cm\": 13.5}]}", "target": "glasses case", "split": "train"}
{"id": "c47168fab2_8", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "c47168fab2_9", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. heater\nB. monitor\nC. banana\nD. laptop", "choices": ["A. heater", "B. monitor", "C. banana", "D. laptop"], "answer": "B. monitor", "target": "keyboard", "split": "train"}
{"id": "c47168fab2_10", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the monitor, about 9 cm away. It is also to the side of and above the heater, about 19 cm away. It is further near the laptop, about 20 cm away.", "target": "keyboard", "split": "train"}
{"id": "c47168fab2_11", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.2}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 18.9}]}", "target": "keyboard", "split": "train"}
{"id": "c47168fab2_12", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "c47168fab2_13", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also under the water jug, about 3 cm away. It is further next to the wet tissue, about 3 cm away.", "target": "mug", "split": "train"}
{"id": "c47168fab2_14", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"water jug\", \"relation\": \"under\", \"distance_cm\": 2.9}, {\"object\": \"wet tissue\", \"relation\": \"next_to\", \"distance_cm\": 3.0}]}", "target": "mug", "split": "train"}
{"id": "c47168fab2_15", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "train"}
{"id": "c47168fab2_16", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. water jug\nB. mug\nC. curtain\nD. wet tissue", "choices": ["A. water jug", "B. mug", "C. curtain", "D. wet tissue"], "answer": "A. water jug", "target": "water bottle", "split": "train"}
{"id": "c47168fab2_17", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also next to the water jug, about 8 cm away. It is further to the side of and above the wet tissue, about 9 cm away.", "target": "water bottle", "split": "train"}
{"id": "c47168fab2_18", "scene_id": "c47168fab2", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"water jug\", \"relation\": \"next_to\", \"distance_cm\": 7.7}, {\"object\": \"wet tissue\", \"relation\": \"side_above\", \"distance_cm\": 9.5}]}", "target": "water bottle", "split": "train"}
{"id": "c49a8c6cff_0", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "c49a8c6cff_1", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. shelf\nC. lamp\nD. books", "choices": ["A. monitor", "B. shelf", "C. lamp", "D. books"], "answer": "D. books", "target": "bottle", "split": "train"}
{"id": "c49a8c6cff_2", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the books, about 5 cm away. It is also to the side of and below the lamp, about 7 cm away. It is further near the monitor, about 39 cm away.", "target": "bottle", "split": "train"}
{"id": "c49a8c6cff_3", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bedside cabinet\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 5.3}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 6.8}]}", "target": "bottle", "split": "train"}
{"id": "c49a8c6cff_4", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "c49a8c6cff_5", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and below the pc, about 15 cm away. It is further near the keyboard, about 16 cm away.", "target": "cup", "split": "train"}
{"id": "c49a8c6cff_6", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pc\", \"relation\": \"side_below\", \"distance_cm\": 15.0}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.6}]}", "target": "cup", "split": "train"}
{"id": "c49a8c6cff_7", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "c49a8c6cff_8", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. pc\nB. paper\nC. postcards\nD. table", "choices": ["A. pc", "B. paper", "C. postcards", "D. table"], "answer": "C. postcards", "target": "headphones", "split": "train"}
{"id": "c49a8c6cff_9", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and below the postcards, about 7 cm away. It is also to the side of and above the table, about 13 cm away. It is further next to the pc, about 14 cm away.", "target": "headphones", "split": "train"}
{"id": "c49a8c6cff_10", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"postcards\", \"relation\": \"side_below\", \"distance_cm\": 7.0}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 12.9}]}", "target": "headphones", "split": "train"}
{"id": "c49a8c6cff_11", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "train"}
{"id": "c49a8c6cff_12", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. monitor\nC. mouse pad\nD. cup", "choices": ["A. mouse", "B. monitor", "C. mouse pad", "D. cup"], "answer": "B. monitor", "target": "keyboard", "split": "train"}
{"id": "c49a8c6cff_13", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the monitor. It is further next to the mouse pad, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "c49a8c6cff_14", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.3}, {\"object\": \"mouse pad\", \"relation\": \"next_to\", \"distance_cm\": 10.0}]}", "target": "keyboard", "split": "train"}
{"id": "c49a8c6cff_15", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "No", "target": "phone charger", "split": "train"}
{"id": "c49a8c6cff_16", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. table\nC. keyboard\nD. notebook", "choices": ["A. monitor", "B. table", "C. keyboard", "D. notebook"], "answer": "B. table", "target": "mouse", "split": "train"}
{"id": "c49a8c6cff_17", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the table, about 4 cm away. It is also to the side of and below the monitor, about 14 cm away. It is further next to the keyboard, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "c49a8c6cff_18", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.8}]}", "target": "mouse", "split": "train"}
{"id": "c49a8c6cff_19", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "c49a8c6cff_20", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. bedside cabinet\nB. office chair\nC. notebook\nD. keyboard", "choices": ["A. bedside cabinet", "B. office chair", "C. notebook", "D. keyboard"], "answer": "A. bedside cabinet", "target": "mouse pad", "split": "train"}
{"id": "c49a8c6cff_21", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the table. It is also to the side of and above the bedside cabinet. It is further next to the keyboard, about 10 cm away.", "target": "mouse pad", "split": "train"}
{"id": "c49a8c6cff_22", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bedside cabinet\", \"relation\": \"side_above\", \"distance_cm\": 2.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 10.0}]}", "target": "mouse pad", "split": "train"}
{"id": "c49a8c6cff_23", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "c49a8c6cff_24", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. office chair\nB. post-it\nC. mouse pad\nD. monitor", "choices": ["A. office chair", "B. post-it", "C. mouse pad", "D. monitor"], "answer": "B. post-it", "target": "notebook", "split": "train"}
{"id": "c49a8c6cff_25", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also next to the post-it, about 5 cm away. It is further to the side of and above the office chair, about 6 cm away.", "target": "notebook", "split": "train"}
{"id": "c49a8c6cff_26", "scene_id": "c49a8c6cff", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"post-it\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 5.8}]}", "target": "notebook", "split": "train"}
{"id": "c4aaedcfd1_0", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "c4aaedcfd1_1", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the desk. It is also below the desk lamp, about 6 cm away. It is further to the side of and above the notebooks, about 7 cm away.", "target": "bottle", "split": "train"}
{"id": "c4aaedcfd1_2", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk lamp\", \"relation\": \"below\", \"distance_cm\": 5.7}, {\"object\": \"notebooks\", \"relation\": \"side_above\", \"distance_cm\": 6.6}]}", "target": "bottle", "split": "train"}
{"id": "c4aaedcfd1_3", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "c4aaedcfd1_4", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the notebooks. It is further under the monitor.", "target": "cup", "split": "train"}
{"id": "c4aaedcfd1_5", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"notebooks\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.4}]}", "target": "cup", "split": "train"}
{"id": "c4aaedcfd1_6", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "c4aaedcfd1_7", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. laptop stand\nC. monitor\nD. office chair", "choices": ["A. mouse", "B. laptop stand", "C. monitor", "D. office chair"], "answer": "C. monitor", "target": "keyboard", "split": "train"}
{"id": "c4aaedcfd1_8", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the monitor, about 4 cm away. It is further next to the laptop stand, about 5 cm away.", "target": "keyboard", "split": "train"}
{"id": "c4aaedcfd1_9", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.7}, {\"object\": \"laptop stand\", \"relation\": \"next_to\", \"distance_cm\": 5.4}]}", "target": "keyboard", "split": "train"}
{"id": "c4aaedcfd1_10", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "c4aaedcfd1_11", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. cup\nC. keyboard\nD. bottle", "choices": ["A. monitor", "B. cup", "C. keyboard", "D. bottle"], "answer": "A. monitor", "target": "mouse", "split": "train"}
{"id": "c4aaedcfd1_12", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also under the monitor, about 3 cm away. It is further next to the keyboard, about 8 cm away.", "target": "mouse", "split": "train"}
{"id": "c4aaedcfd1_13", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.8}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.4}]}", "target": "mouse", "split": "train"}
{"id": "c4aaedcfd1_14", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "c4aaedcfd1_15", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor\nB. office chair\nC. router\nD. keyboard", "choices": ["A. monitor", "B. office chair", "C. router", "D. keyboard"], "answer": "A. monitor", "target": "power strip", "split": "train"}
{"id": "c4aaedcfd1_16", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the desk. It is also to the side of and below the monitor, about 26 cm away. It is further near the keyboard, about 34 cm away.", "target": "power strip", "split": "train"}
{"id": "c4aaedcfd1_17", "scene_id": "c4aaedcfd1", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 25.6}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 34.2}]}", "target": "power strip", "split": "train"}
{"id": "c4d4cb61f6_0", "scene_id": "c4d4cb61f6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "c4d4cb61f6_1", "scene_id": "c4d4cb61f6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the lotion bottle in this scene?\nA. storage basket\nB. curtain\nC. table lamp\nD. pillow", "choices": ["A. storage basket", "B. curtain", "C. table lamp", "D. pillow"], "answer": "C. table lamp", "target": "lotion bottle", "split": "train"}
{"id": "c4d4cb61f6_2", "scene_id": "c4d4cb61f6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the lotion bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The lotion bottle is next to the table lamp, about 3 cm away. It is also near the storage basket, about 18 cm away. It is further to the side of and above the pillow, about 44 cm away.", "target": "lotion bottle", "split": "train"}
{"id": "c4d4cb61f6_3", "scene_id": "c4d4cb61f6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the lotion bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"lotion bottle\", \"support_surface\": \"bedside cabinet\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"next_to\", \"distance_cm\": 2.7}, {\"object\": \"storage basket\", \"relation\": \"near\", \"distance_cm\": 18.0}]}", "target": "lotion bottle", "split": "train"}
{"id": "c4ed7b5ebe_0", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "c4ed7b5ebe_1", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. food container\nB. laptop bag\nC. paper\nD. shelf", "choices": ["A. food container", "B. laptop bag", "C. paper", "D. shelf"], "answer": "B. laptop bag", "target": "bottle", "split": "train"}
{"id": "c4ed7b5ebe_2", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the desk. It is also next to the laptop bag, about 9 cm away. It is further to the side of and above the paper, about 15 cm away.", "target": "bottle", "split": "train"}
{"id": "c4ed7b5ebe_3", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop bag\", \"relation\": \"next_to\", \"distance_cm\": 9.4}, {\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 14.7}]}", "target": "bottle", "split": "train"}
{"id": "c4ed7b5ebe_4", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "Yes", "target": "toothbrush", "split": "train"}
{"id": "c4ed7b5ebe_5", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. sink\nB. faucet\nC. bag\nD. bottle", "choices": ["A. sink", "B. faucet", "C. bag", "D. bottle"], "answer": "C. bag", "target": "toothbrush", "split": "train"}
{"id": "c4ed7b5ebe_6", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is on the shelf. It is also next to the bag, about 3 cm away. It is further next to the bottle, about 6 cm away.", "target": "toothbrush", "split": "train"}
{"id": "c4ed7b5ebe_7", "scene_id": "c4ed7b5ebe", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 2.8}, {\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 6.5}]}", "target": "toothbrush", "split": "train"}
{"id": "c50d2d1d42_0", "scene_id": "c50d2d1d42", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "c50d2d1d42_1", "scene_id": "c50d2d1d42", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the mug, about 9 cm away. It is also to the side of and below the plastic bottle, about 9 cm away. It is further to the side of and above the office chair, about 17 cm away.", "target": "headphones", "split": "train"}
{"id": "c50d2d1d42_2", "scene_id": "c50d2d1d42", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 8.8}, {\"object\": \"plastic bottle\", \"relation\": \"side_below\", \"distance_cm\": 9.1}]}", "target": "headphones", "split": "train"}
{"id": "c50d2d1d42_3", "scene_id": "c50d2d1d42", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "c50d2d1d42_4", "scene_id": "c50d2d1d42", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. monitor\nB. coaster\nC. food container\nD. keyboard", "choices": ["A. monitor", "B. coaster", "C. food container", "D. keyboard"], "answer": "B. coaster", "target": "mousepad", "split": "train"}
{"id": "c50d2d1d42_5", "scene_id": "c50d2d1d42", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is next to the coaster, about 4 cm away. It is also next to the keyboard, about 6 cm away. It is further to the side of and below the monitor, about 8 cm away.", "target": "mousepad", "split": "train"}
{"id": "c50d2d1d42_6", "scene_id": "c50d2d1d42", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"coaster\", \"relation\": \"next_to\", \"distance_cm\": 3.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.2}]}", "target": "mousepad", "split": "train"}
{"id": "c5439f4607_0", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "c5439f4607_1", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap in this scene?\nA. laboratory equipment\nB. first aid cabinet\nC. sink pipe\nD. faucet", "choices": ["A. laboratory equipment", "B. first aid cabinet", "C. sink pipe", "D. faucet"], "answer": "A. laboratory equipment", "target": "soap", "split": "train"}
{"id": "c5439f4607_2", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap is to the side of and below the laboratory equipment, about 10 cm away. It is also to the side of and below the faucet, about 20 cm away. It is further to the side of and above the sink pipe, about 28 cm away.", "target": "soap", "split": "train"}
{"id": "c5439f4607_3", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap\", \"support_surface\": \"first aid cabinet\", \"references\": [{\"object\": \"laboratory equipment\", \"relation\": \"side_below\", \"distance_cm\": 10.2}, {\"object\": \"faucet\", \"relation\": \"side_below\", \"distance_cm\": 19.8}]}", "target": "soap", "split": "train"}
{"id": "c5439f4607_4", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "c5439f4607_5", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. outlet\nB. laboratory equipment\nC. faucet\nD. water pipe", "choices": ["A. outlet", "B. laboratory equipment", "C. faucet", "D. water pipe"], "answer": "D. water pipe", "target": "spray bottle", "split": "train"}
{"id": "c5439f4607_6", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and below the water pipe, about 13 cm away. It is also to the side of and above the outlet, about 41 cm away. It is further to the side of and above the laboratory equipment, about 59 cm away.", "target": "spray bottle", "split": "train"}
{"id": "c5439f4607_7", "scene_id": "c5439f4607", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"first aid cabinet\", \"references\": [{\"object\": \"water pipe\", \"relation\": \"side_below\", \"distance_cm\": 13.0}, {\"object\": \"outlet\", \"relation\": \"side_above\", \"distance_cm\": 41.5}]}", "target": "spray bottle", "split": "train"}
{"id": "c545851c4f_0", "scene_id": "c545851c4f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "c545851c4f_1", "scene_id": "c545851c4f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. cabinet\nB. laptop\nC. partition\nD. lab machine", "choices": ["A. cabinet", "B. laptop", "C. partition", "D. lab machine"], "answer": "A. cabinet", "target": "mouse", "split": "train"}
{"id": "c545851c4f_2", "scene_id": "c545851c4f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the cabinet, about 4 cm away. It is also to the side of and below the lab machine, about 7 cm away. It is further to the side of and below the laptop, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "c545851c4f_3", "scene_id": "c545851c4f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"lab machine\", \"relation\": \"side_below\", \"distance_cm\": 7.1}]}", "target": "mouse", "split": "train"}
{"id": "c6c184cec0_0", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "c6c184cec0_1", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the hairbrush in this scene?\nA. shelf\nB. laptop\nC. backpack\nD. plastic bottle", "choices": ["A. shelf", "B. laptop", "C. backpack", "D. plastic bottle"], "answer": "B. laptop", "target": "hairbrush", "split": "train"}
{"id": "c6c184cec0_2", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the hairbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The hairbrush is next to the laptop. It is also next to the plastic bottle, about 5 cm away. It is further to the side of and above the shelf, about 8 cm away.", "target": "hairbrush", "split": "train"}
{"id": "c6c184cec0_3", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the hairbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"hairbrush\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 2.2}, {\"object\": \"plastic bottle\", \"relation\": \"next_to\", \"distance_cm\": 4.5}]}", "target": "hairbrush", "split": "train"}
{"id": "c6c184cec0_4", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "c6c184cec0_5", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. wall lamp\nB. trash can\nC. plastic bottle\nD. socket", "choices": ["A. wall lamp", "B. trash can", "C. plastic bottle", "D. socket"], "answer": "C. plastic bottle", "target": "mouse", "split": "train"}
{"id": "c6c184cec0_6", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and below the plastic bottle, about 16 cm away. It is also to the side of and above the socket, about 34 cm away. It is further to the side of and above the trash can, about 52 cm away.", "target": "mouse", "split": "train"}
{"id": "c6c184cec0_7", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"wall lamp\", \"references\": [{\"object\": \"plastic bottle\", \"relation\": \"side_below\", \"distance_cm\": 16.4}, {\"object\": \"socket\", \"relation\": \"side_above\", \"distance_cm\": 34.3}]}", "target": "mouse", "split": "train"}
{"id": "c6c184cec0_8", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "train"}
{"id": "c6c184cec0_9", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. tv\nB. luggage\nC. shelf\nD. clothes", "choices": ["A. tv", "B. luggage", "C. shelf", "D. clothes"], "answer": "C. shelf", "target": "remote", "split": "train"}
{"id": "c6c184cec0_10", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is to the side of and above the shelf, about 7 cm away. It is also to the side of and below the tv, about 18 cm away. It is further to the side of and above the luggage, about 54 cm away.", "target": "remote", "split": "train"}
{"id": "c6c184cec0_11", "scene_id": "c6c184cec0", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 7.4}, {\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 18.0}]}", "target": "remote", "split": "train"}
{"id": "c768a82cd0_0", "scene_id": "c768a82cd0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "c768a82cd0_1", "scene_id": "c768a82cd0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. door frame\nB. light switch\nC. oven\nD. kitchen apron", "choices": ["A. door frame", "B. light switch", "C. oven", "D. kitchen apron"], "answer": "A. door frame", "target": "tote bag", "split": "train"}
{"id": "c768a82cd0_2", "scene_id": "c768a82cd0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is attached to the door. It is also to the side of and above the door frame, about 5 cm away. It is further above the kitchen apron, about 7 cm away.", "target": "tote bag", "split": "train"}
{"id": "c768a82cd0_3", "scene_id": "c768a82cd0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door frame\", \"relation\": \"side_above\", \"distance_cm\": 5.4}, {\"object\": \"kitchen apron\", \"relation\": \"above\", \"distance_cm\": 7.2}]}", "target": "tote bag", "split": "train"}
{"id": "c842edbdf5_0", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "c842edbdf5_1", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. telephone\nC. cabinet\nD. mouse", "choices": ["A. monitor", "B. telephone", "C. cabinet", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "train"}
{"id": "c842edbdf5_2", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 12 cm away. It is further to the side of and above the cabinet, about 14 cm away.", "target": "keyboard", "split": "train"}
{"id": "c842edbdf5_3", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 11.8}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 14.4}]}", "target": "keyboard", "split": "train"}
{"id": "c842edbdf5_4", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the comb appear?", "choices": null, "answer": "No", "target": "comb", "split": "train"}
{"id": "c842edbdf5_5", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. book\nB. books\nC. office chair\nD. keyboard", "choices": ["A. book", "B. books", "C. office chair", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "c842edbdf5_6", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 12 cm away. It is further near the books, about 22 cm away.", "target": "mouse", "split": "train"}
{"id": "c842edbdf5_7", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 11.8}, {\"object\": \"books\", \"relation\": \"near\", \"distance_cm\": 22.3}]}", "target": "mouse", "split": "train"}
{"id": "c842edbdf5_8", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "c842edbdf5_9", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. books\nB. cabinet\nC. mouse\nD. book", "choices": ["A. books", "B. cabinet", "C. mouse", "D. book"], "answer": "D. book", "target": "pen holder", "split": "train"}
{"id": "c842edbdf5_10", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also to the side of and above the book, about 22 cm away. It is further to the side of and above the books, about 24 cm away.", "target": "pen holder", "split": "train"}
{"id": "c842edbdf5_11", "scene_id": "c842edbdf5", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 21.8}, {\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 24.5}]}", "target": "pen holder", "split": "train"}
{"id": "c8eeef6427_0", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "Yes", "target": "charger", "split": "train"}
{"id": "c8eeef6427_1", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. lan port\nB. paper\nC. power extender\nD. power socket", "choices": ["A. lan port", "B. paper", "C. power extender", "D. power socket"], "answer": "C. power extender", "target": "charger", "split": "train"}
{"id": "c8eeef6427_2", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is next to the power extender, about 14 cm away. It is also to the side of and below the lan port, about 22 cm away. It is further to the side of and below the power socket, about 42 cm away.", "target": "charger", "split": "train"}
{"id": "c8eeef6427_3", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"paper\", \"references\": [{\"object\": \"power extender\", \"relation\": \"next_to\", \"distance_cm\": 14.0}, {\"object\": \"lan port\", \"relation\": \"side_below\", \"distance_cm\": 21.8}]}", "target": "charger", "split": "train"}
{"id": "c8eeef6427_4", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "No", "target": "phone charger", "split": "train"}
{"id": "c8eeef6427_5", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. curtain\nB. table lamp\nC. blinds roller\nD. paper", "choices": ["A. curtain", "B. table lamp", "C. blinds roller", "D. paper"], "answer": "B. table lamp", "target": "water bottle", "split": "train"}
{"id": "c8eeef6427_6", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also next to the table lamp, about 3 cm away. It is further to the side of and above the paper, about 9 cm away.", "target": "water bottle", "split": "train"}
{"id": "c8eeef6427_7", "scene_id": "c8eeef6427", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 9.4}]}", "target": "water bottle", "split": "train"}
{"id": "c8fc01c453_0", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "c8fc01c453_1", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. shoes\nB. laundry basket\nC. luggage\nD. cloth", "choices": ["A. shoes", "B. laundry basket", "C. luggage", "D. cloth"], "answer": "C. luggage", "target": "backpack", "split": "train"}
{"id": "c8fc01c453_2", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the luggage, about 7 cm away. It is also to the side of and above the shoes, about 21 cm away. It is further near the cloth, about 21 cm away.", "target": "backpack", "split": "train"}
{"id": "c8fc01c453_3", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"laundry basket\", \"references\": [{\"object\": \"luggage\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 21.0}]}", "target": "backpack", "split": "train"}
{"id": "c8fc01c453_4", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "c8fc01c453_5", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. bottle\nB. bags\nC. luggage\nD. box", "choices": ["A. bottle", "B. bags", "C. luggage", "D. box"], "answer": "A. bottle", "target": "spray bottle", "split": "train"}
{"id": "c8fc01c453_6", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is next to the bottle, about 7 cm away. It is also to the side of and below the box, about 16 cm away. It is further to the side of and above the bags, about 24 cm away.", "target": "spray bottle", "split": "train"}
{"id": "c8fc01c453_7", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 16.1}]}", "target": "spray bottle", "split": "train"}
{"id": "c8fc01c453_8", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "c8fc01c453_9", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. tray\nB. bottle\nC. clothes iron\nD. box", "choices": ["A. tray", "B. bottle", "C. clothes iron", "D. box"], "answer": "D. box", "target": "tote bag", "split": "train"}
{"id": "c8fc01c453_10", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is on the shelf. It is also under the box, about 3 cm away. It is further under the tray, about 4 cm away.", "target": "tote bag", "split": "train"}
{"id": "c8fc01c453_11", "scene_id": "c8fc01c453", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"under\", \"distance_cm\": 2.8}, {\"object\": \"tray\", \"relation\": \"under\", \"distance_cm\": 4.4}]}", "target": "tote bag", "split": "train"}
{"id": "c9a8357e8f_0", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "c9a8357e8f_1", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. power sockets\nB. cable duct\nC. tablet\nD. book", "choices": ["A. power sockets", "B. cable duct", "C. tablet", "D. book"], "answer": "C. tablet", "target": "backpack", "split": "train"}
{"id": "c9a8357e8f_2", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and above the tablet, about 7 cm away. It is further to the side of and below the cable duct, about 8 cm away.", "target": "backpack", "split": "train"}
{"id": "c9a8357e8f_3", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"tablet\", \"relation\": \"side_above\", \"distance_cm\": 6.6}, {\"object\": \"cable duct\", \"relation\": \"side_below\", \"distance_cm\": 8.0}]}", "target": "backpack", "split": "train"}
{"id": "c9a8357e8f_4", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "c9a8357e8f_5", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. light switch\nB. door window\nC. book\nD. backpack", "choices": ["A. light switch", "B. door window", "C. book", "D. backpack"], "answer": "B. door window", "target": "phone charger", "split": "train"}
{"id": "c9a8357e8f_6", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the door, about 10 cm away. It is also to the side of and below the door window, about 29 cm away. It is further below the light switch, about 73 cm away.", "target": "phone charger", "split": "train"}
{"id": "c9a8357e8f_7", "scene_id": "c9a8357e8f", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 10.1}, {\"object\": \"door window\", \"relation\": \"side_below\", \"distance_cm\": 29.2}]}", "target": "phone charger", "split": "train"}
{"id": "ca3b651c23_0", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "No", "target": "highlighter", "split": "train"}
{"id": "ca3b651c23_1", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. dustbin\nB. dish sponges\nC. power sockets\nD. paper towel", "choices": ["A. dustbin", "B. dish sponges", "C. power sockets", "D. paper towel"], "answer": "B. dish sponges", "target": "bottle", "split": "train"}
{"id": "ca3b651c23_2", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the dish sponges, about 10 cm away. It is also to the side of and above the dustbin, about 39 cm away. It is further to the side of and below the paper towel, about 43 cm away.", "target": "bottle", "split": "train"}
{"id": "ca3b651c23_3", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"window\", \"references\": [{\"object\": \"dish sponges\", \"relation\": \"side_above\", \"distance_cm\": 10.2}, {\"object\": \"dustbin\", \"relation\": \"side_above\", \"distance_cm\": 39.0}]}", "target": "bottle", "split": "train"}
{"id": "ca3b651c23_4", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "ca3b651c23_5", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the table, about 3 cm away. It is also to the side of and above the mug, about 4 cm away. It is further to the side of and above the chair, about 19 cm away.", "target": "cup", "split": "train"}
{"id": "ca3b651c23_6", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 3.1}, {\"object\": \"mug\", \"relation\": \"side_above\", \"distance_cm\": 3.7}]}", "target": "cup", "split": "train"}
{"id": "ca3b651c23_7", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "Yes", "target": "game controller", "split": "train"}
{"id": "ca3b651c23_8", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the game controller in this scene?\nA. keyboard\nB. mouse\nC. monitor\nD. toilet paper roll", "choices": ["A. keyboard", "B. mouse", "C. monitor", "D. toilet paper roll"], "answer": "A. keyboard", "target": "game controller", "split": "train"}
{"id": "ca3b651c23_9", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the game controller in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The game controller is on the table. It is also next to the keyboard. It is further next to the toilet paper roll, about 8 cm away.", "target": "game controller", "split": "train"}
{"id": "ca3b651c23_10", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the game controller in structured JSON format.", "choices": null, "answer": "{\"target\": \"game controller\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"toilet paper roll\", \"relation\": \"next_to\", \"distance_cm\": 8.1}]}", "target": "game controller", "split": "train"}
{"id": "ca3b651c23_11", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "ca3b651c23_12", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. toilet paper roll\nB. game controller\nC. headphones\nD. keyboard", "choices": ["A. toilet paper roll", "B. game controller", "C. headphones", "D. keyboard"], "answer": "C. headphones", "target": "mouse", "split": "train"}
{"id": "ca3b651c23_13", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the headphones, about 8 cm away. It is further next to the toilet paper roll, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "ca3b651c23_14", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 7.8}, {\"object\": \"toilet paper roll\", \"relation\": \"next_to\", \"distance_cm\": 10.2}]}", "target": "mouse", "split": "train"}
{"id": "ca3b651c23_15", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "ca3b651c23_16", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. chair\nB. keyboard\nC. monitor\nD. cup", "choices": ["A. chair", "B. keyboard", "C. monitor", "D. cup"], "answer": "D. cup", "target": "mug", "split": "train"}
{"id": "ca3b651c23_17", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also under the cup, about 4 cm away. It is further to the side of and below the monitor, about 22 cm away.", "target": "mug", "split": "train"}
{"id": "ca3b651c23_18", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cup\", \"relation\": \"under\", \"distance_cm\": 3.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 21.8}]}", "target": "mug", "split": "train"}
{"id": "ca3b651c23_19", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "Yes", "target": "pencil holder", "split": "train"}
{"id": "ca3b651c23_20", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil holder in this scene?\nA. headphones\nB. game controller\nC. nintendo switch\nD. monitor", "choices": ["A. headphones", "B. game controller", "C. nintendo switch", "D. monitor"], "answer": "C. nintendo switch", "target": "pencil holder", "split": "train"}
{"id": "ca3b651c23_21", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil holder is on the table. It is also next to the nintendo switch, about 5 cm away. It is further to the side of and below the monitor, about 7 cm away.", "target": "pencil holder", "split": "train"}
{"id": "ca3b651c23_22", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"nintendo switch\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.3}]}", "target": "pencil holder", "split": "train"}
{"id": "ca3b651c23_23", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "ca3b651c23_24", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. laptop stand\nB. coffee machine\nC. dustbin\nD. kettle", "choices": ["A. laptop stand", "B. coffee machine", "C. dustbin", "D. kettle"], "answer": "C. dustbin", "target": "power strip", "split": "train"}
{"id": "ca3b651c23_25", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the dustbin, about 35 cm away. It is also to the side of and below the laptop stand, about 64 cm away. It is further to the side of and below the coffee machine, about 66 cm away.", "target": "power strip", "split": "train"}
{"id": "ca3b651c23_26", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"dustbin\", \"relation\": \"side_below\", \"distance_cm\": 35.0}, {\"object\": \"laptop stand\", \"relation\": \"side_below\", \"distance_cm\": 64.1}]}", "target": "power strip", "split": "train"}
{"id": "ca3b651c23_27", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "ca3b651c23_28", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. step stool\nB. food tray\nC. plate\nD. clothes", "choices": ["A. step stool", "B. food tray", "C. plate", "D. clothes"], "answer": "A. step stool", "target": "tissue box", "split": "train"}
{"id": "ca3b651c23_29", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the table. It is also to the side of and above the step stool, about 10 cm away. It is further to the side of and above the plate, about 32 cm away.", "target": "tissue box", "split": "train"}
{"id": "ca3b651c23_30", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"step stool\", \"relation\": \"side_above\", \"distance_cm\": 9.9}, {\"object\": \"plate\", \"relation\": \"side_above\", \"distance_cm\": 31.7}]}", "target": "tissue box", "split": "train"}
{"id": "ca3b651c23_31", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "ca3b651c23_32", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. kitchen cabinet\nB. spatula\nC. peeler\nD. mugs", "choices": ["A. kitchen cabinet", "B. spatula", "C. peeler", "D. mugs"], "answer": "D. mugs", "target": "water bottle", "split": "train"}
{"id": "ca3b651c23_33", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "33", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is to the side of and above the mugs, about 10 cm away. It is also to the side of and above the kitchen cabinet, about 12 cm away. It is further to the side of and below the spatula, about 16 cm away.", "target": "water bottle", "split": "train"}
{"id": "ca3b651c23_34", "scene_id": "ca3b651c23", "source_dataset": "scannetpp", "local_sample_id": "34", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"mugs\", \"relation\": \"side_above\", \"distance_cm\": 10.2}, {\"object\": \"kitchen cabinet\", \"relation\": \"side_above\", \"distance_cm\": 11.8}]}", "target": "water bottle", "split": "train"}
{"id": "cab239278a_0", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "cab239278a_1", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the water bottle, about 5 cm away. It is also next to the chair, about 6 cm away. It is further to the side of and below the whiteboard markers, about 58 cm away.", "target": "keyboard", "split": "train"}
{"id": "cab239278a_2", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"water bottle\", \"relation\": \"side_below\", \"distance_cm\": 5.5}, {\"object\": \"chair\", \"relation\": \"next_to\", \"distance_cm\": 6.4}]}", "target": "keyboard", "split": "train"}
{"id": "cab239278a_3", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "cab239278a_4", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. flipchart\nB. whiteboard markers\nC. chair\nD. mini pc", "choices": ["A. flipchart", "B. whiteboard markers", "C. chair", "D. mini pc"], "answer": "D. mini pc", "target": "mouse", "split": "train"}
{"id": "cab239278a_5", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the mini pc, about 15 cm away. It is also to the side of and above the chair, about 40 cm away. It is further to the side of and below the whiteboard markers, about 95 cm away.", "target": "mouse", "split": "train"}
{"id": "cab239278a_6", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mini pc\", \"relation\": \"next_to\", \"distance_cm\": 14.8}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 40.5}]}", "target": "mouse", "split": "train"}
{"id": "cab239278a_7", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "cab239278a_8", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. keyboard\nB. flipchart\nC. chair\nD. mini pc", "choices": ["A. keyboard", "B. flipchart", "C. chair", "D. mini pc"], "answer": "A. keyboard", "target": "water bottle", "split": "train"}
{"id": "cab239278a_9", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also to the side of and above the keyboard, about 5 cm away. It is further to the side of and above the mini pc, about 22 cm away.", "target": "water bottle", "split": "train"}
{"id": "cab239278a_10", "scene_id": "cab239278a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 5.5}, {\"object\": \"mini pc\", \"relation\": \"side_above\", \"distance_cm\": 21.6}]}", "target": "water bottle", "split": "train"}
{"id": "cb7785f6ad_0", "scene_id": "cb7785f6ad", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "cb7785f6ad_1", "scene_id": "cb7785f6ad", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. kettle\nB. mixer\nC. windowsill\nD. kitchen paper", "choices": ["A. kettle", "B. mixer", "C. windowsill", "D. kitchen paper"], "answer": "D. kitchen paper", "target": "mug", "split": "train"}
{"id": "cb7785f6ad_2", "scene_id": "cb7785f6ad", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is to the side of and below the kitchen paper, about 8 cm away. It is also to the side of and below the kettle, about 17 cm away. It is further to the side of and below the mixer, about 31 cm away.", "target": "mug", "split": "train"}
{"id": "cb7785f6ad_3", "scene_id": "cb7785f6ad", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"kitchen paper\", \"relation\": \"side_below\", \"distance_cm\": 7.6}, {\"object\": \"kettle\", \"relation\": \"side_below\", \"distance_cm\": 17.1}]}", "target": "mug", "split": "train"}
{"id": "cba701332a_0", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "cba701332a_1", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. oil diffuser\nB. toilet paper\nC. box\nD. power extender", "choices": ["A. oil diffuser", "B. toilet paper", "C. box", "D. power extender"], "answer": "B. toilet paper", "target": "bottle", "split": "train"}
{"id": "cba701332a_2", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also near the toilet paper, about 22 cm away. It is further to the side of and above the box, about 29 cm away.", "target": "bottle", "split": "train"}
{"id": "cba701332a_3", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"near\", \"distance_cm\": 21.5}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 29.0}]}", "target": "bottle", "split": "train"}
{"id": "cba701332a_4", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "cba701332a_5", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the calculator in this scene?\nA. chair\nB. keyboard\nC. book\nD. papers", "choices": ["A. chair", "B. keyboard", "C. book", "D. papers"], "answer": "C. book", "target": "calculator", "split": "train"}
{"id": "cba701332a_6", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the calculator in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The calculator is on the table. It is also next to the book, about 4 cm away. It is further next to the keyboard, about 6 cm away.", "target": "calculator", "split": "train"}
{"id": "cba701332a_7", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the calculator in structured JSON format.", "choices": null, "answer": "{\"target\": \"calculator\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.2}]}", "target": "calculator", "split": "train"}
{"id": "cba701332a_8", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "cba701332a_9", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. file holder\nB. keyboard\nC. chair\nD. backpack", "choices": ["A. file holder", "B. keyboard", "C. chair", "D. backpack"], "answer": "A. file holder", "target": "headphones", "split": "train"}
{"id": "cba701332a_10", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also next to the file holder, about 6 cm away. It is further to the side of and above the chair, about 15 cm away.", "target": "headphones", "split": "train"}
{"id": "cba701332a_11", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"file holder\", \"relation\": \"next_to\", \"distance_cm\": 5.7}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 15.3}]}", "target": "headphones", "split": "train"}
{"id": "cba701332a_12", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "cba701332a_13", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. calculator\nB. power extender\nC. book\nD. headphones", "choices": ["A. calculator", "B. power extender", "C. book", "D. headphones"], "answer": "A. calculator", "target": "keyboard", "split": "train"}
{"id": "cba701332a_14", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the calculator, about 6 cm away. It is further next to the book, about 12 cm away.", "target": "keyboard", "split": "train"}
{"id": "cba701332a_15", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"calculator\", \"relation\": \"next_to\", \"distance_cm\": 6.2}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 11.8}]}", "target": "keyboard", "split": "train"}
{"id": "cba701332a_16", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "cba701332a_17", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. power extender\nB. papers\nC. laptop stand\nD. laptop bag", "choices": ["A. power extender", "B. papers", "C. laptop stand", "D. laptop bag"], "answer": "C. laptop stand", "target": "pen", "split": "train"}
{"id": "cba701332a_18", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the table. It is also next to the laptop stand, about 4 cm away. It is further next to the power extender, about 12 cm away.", "target": "pen", "split": "train"}
{"id": "cba701332a_19", "scene_id": "cba701332a", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop stand\", \"relation\": \"next_to\", \"distance_cm\": 3.9}, {\"object\": \"power extender\", \"relation\": \"next_to\", \"distance_cm\": 11.5}]}", "target": "pen", "split": "train"}
{"id": "cc0aa81452_0", "scene_id": "cc0aa81452", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "cc0aa81452_1", "scene_id": "cc0aa81452", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. long brush\nB. long pan\nC. sink\nD. sign", "choices": ["A. long brush", "B. long pan", "C. sink", "D. sign"], "answer": "D. sign", "target": "spray bottle", "split": "train"}
{"id": "cc0aa81452_2", "scene_id": "cc0aa81452", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the table. It is also near the sign, about 17 cm away. It is further to the side of and above the long brush, about 27 cm away.", "target": "spray bottle", "split": "train"}
{"id": "cc0aa81452_3", "scene_id": "cc0aa81452", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"sign\", \"relation\": \"near\", \"distance_cm\": 17.3}, {\"object\": \"long brush\", \"relation\": \"side_above\", \"distance_cm\": 26.9}]}", "target": "spray bottle", "split": "train"}
{"id": "cc0aa81452_4", "scene_id": "cc0aa81452", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "Yes", "target": "tv remote", "split": "train"}
{"id": "cc5237fd77_0", "scene_id": "cc5237fd77", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "cc5237fd77_1", "scene_id": "cc5237fd77", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. cabinet\nC. computer tower\nD. monitor", "choices": ["A. chair", "B. cabinet", "C. computer tower", "D. monitor"], "answer": "B. cabinet", "target": "keyboard", "split": "train"}
{"id": "cc5237fd77_2", "scene_id": "cc5237fd77", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the cabinet, about 9 cm away. It is further to the side of and below the computer tower, about 12 cm away.", "target": "keyboard", "split": "train"}
{"id": "cc5237fd77_3", "scene_id": "cc5237fd77", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 9.1}, {\"object\": \"computer tower\", \"relation\": \"side_below\", \"distance_cm\": 11.5}]}", "target": "keyboard", "split": "train"}
{"id": "cc5ea8026c_0", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "cc5ea8026c_1", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. toilet\nB. toilet flush button\nC. tube\nD. toliet paper", "choices": ["A. toilet", "B. toilet flush button", "C. tube", "D. toliet paper"], "answer": "C. tube", "target": "bottle", "split": "train"}
{"id": "cc5ea8026c_2", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also to the side of and above the toilet flush button, about 17 cm away. It is further near the toliet paper, about 32 cm away.", "target": "bottle", "split": "train"}
{"id": "cc5ea8026c_3", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"toilet flush button\", \"relation\": \"side_above\", \"distance_cm\": 16.8}, {\"object\": \"toliet paper\", \"relation\": \"near\", \"distance_cm\": 31.5}]}", "target": "bottle", "split": "train"}
{"id": "cc5ea8026c_4", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "Yes", "target": "soap", "split": "train"}
{"id": "cc5ea8026c_5", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap in this scene?\nA. power socket\nB. tap\nC. soap dish\nD. towel rod", "choices": ["A. power socket", "B. tap", "C. soap dish", "D. towel rod"], "answer": "D. towel rod", "target": "soap", "split": "train"}
{"id": "cc5ea8026c_6", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap is next to the towel rod, about 13 cm away. It is also to the side of and below the tap, about 14 cm away. It is further to the side of and below the power socket, about 16 cm away.", "target": "soap", "split": "train"}
{"id": "cc5ea8026c_7", "scene_id": "cc5ea8026c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap\", \"support_surface\": \"sink\", \"references\": [{\"object\": \"towel rod\", \"relation\": \"next_to\", \"distance_cm\": 12.8}, {\"object\": \"tap\", \"relation\": \"side_below\", \"distance_cm\": 14.3}]}", "target": "soap", "split": "train"}
{"id": "ccfd3ed9c7_0", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "ccfd3ed9c7_1", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. computer tower\nB. cabinet\nC. box\nD. trash can", "choices": ["A. computer tower", "B. cabinet", "C. box", "D. trash can"], "answer": "D. trash can", "target": "cable", "split": "train"}
{"id": "ccfd3ed9c7_2", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and below the trash can, about 23 cm away.", "target": "cable", "split": "train"}
{"id": "ccfd3ed9c7_3", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"trash can\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 23.2}]}", "target": "cable", "split": "train"}
{"id": "ccfd3ed9c7_4", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "ccfd3ed9c7_5", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. mug\nB. keyboard\nC. cabinet\nD. box", "choices": ["A. mug", "B. keyboard", "C. cabinet", "D. box"], "answer": "A. mug", "target": "cup", "split": "train"}
{"id": "ccfd3ed9c7_6", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also next to the mug, about 11 cm away. It is further near the keyboard, about 15 cm away.", "target": "cup", "split": "train"}
{"id": "ccfd3ed9c7_7", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 11.0}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.1}]}", "target": "cup", "split": "train"}
{"id": "ccfd3ed9c7_8", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "ccfd3ed9c7_9", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. office chair\nC. mouse\nD. headphone", "choices": ["A. monitor", "B. office chair", "C. mouse", "D. headphone"], "answer": "D. headphone", "target": "keyboard", "split": "train"}
{"id": "ccfd3ed9c7_10", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the headphone, about 6 cm away. It is further next to the mouse, about 11 cm away.", "target": "keyboard", "split": "train"}
{"id": "ccfd3ed9c7_11", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headphone\", \"relation\": \"next_to\", \"distance_cm\": 5.6}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 10.9}]}", "target": "keyboard", "split": "train"}
{"id": "ccfd3ed9c7_12", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "ccfd3ed9c7_13", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. keyboard\nC. mug\nD. monitor", "choices": ["A. office chair", "B. keyboard", "C. mug", "D. monitor"], "answer": "D. monitor", "target": "mouse", "split": "train"}
{"id": "ccfd3ed9c7_14", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also under the monitor, about 5 cm away. It is further next to the keyboard, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "ccfd3ed9c7_15", "scene_id": "ccfd3ed9c7", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 5.0}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 10.1}]}", "target": "mouse", "split": "train"}
{"id": "cd2e0629fe_0", "scene_id": "cd2e0629fe", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "cd2e0629fe_1", "scene_id": "cd2e0629fe", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. food products\nB. jar\nC. box\nD. basket", "choices": ["A. food products", "B. jar", "C. box", "D. basket"], "answer": "A. food products", "target": "cup", "split": "train"}
{"id": "cd2e0629fe_2", "scene_id": "cd2e0629fe", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and below the basket, about 22 cm away. It is also to the side of and below the jar, about 29 cm away. It is further to the side of and above the wall light, about 38 cm away.", "target": "cup", "split": "train"}
{"id": "cd2e0629fe_3", "scene_id": "cd2e0629fe", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"open cabinet\", \"references\": [{\"object\": \"basket\", \"relation\": \"side_below\", \"distance_cm\": 21.8}, {\"object\": \"jar\", \"relation\": \"side_below\", \"distance_cm\": 29.5}]}", "target": "cup", "split": "train"}
{"id": "ce3a7945ce_0", "scene_id": "ce3a7945ce", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "ce3a7945ce_1", "scene_id": "ce3a7945ce", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "ce3a7945ce_2", "scene_id": "ce3a7945ce", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. trolley table\nB. router\nC. bottle\nD. cabinet", "choices": ["A. trolley table", "B. router", "C. bottle", "D. cabinet"], "answer": "B. router", "target": "power strip", "split": "train"}
{"id": "ce3a7945ce_3", "scene_id": "ce3a7945ce", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the cabinet. It is also near the router, about 46 cm away.", "target": "power strip", "split": "train"}
{"id": "ce3a7945ce_4", "scene_id": "ce3a7945ce", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"router\", \"relation\": \"near\", \"distance_cm\": 45.7}]}", "target": "power strip", "split": "train"}
{"id": "cec8312f4e_0", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "cec8312f4e_1", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. suitcase\nB. monitor\nC. desk\nD. stool", "choices": ["A. suitcase", "B. monitor", "C. desk", "D. stool"], "answer": "D. stool", "target": "keyboard", "split": "train"}
{"id": "cec8312f4e_2", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the stool, about 8 cm away. It is also next to the desk, about 13 cm away. It is further to the side of and below the monitor, about 32 cm away.", "target": "keyboard", "split": "train"}
{"id": "cec8312f4e_3", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"stool\", \"relation\": \"side_above\", \"distance_cm\": 7.8}, {\"object\": \"desk\", \"relation\": \"next_to\", \"distance_cm\": 13.4}]}", "target": "keyboard", "split": "train"}
{"id": "cec8312f4e_4", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "cec8312f4e_5", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. books\nB. monitor\nC. newspaper\nD. table", "choices": ["A. books", "B. monitor", "C. newspaper", "D. table"], "answer": "D. table", "target": "mouse", "split": "train"}
{"id": "cec8312f4e_6", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the table, about 5 cm away. It is also to the side of and below the monitor, about 16 cm away. It is further near the newspaper, about 38 cm away.", "target": "mouse", "split": "train"}
{"id": "cec8312f4e_7", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 15.6}]}", "target": "mouse", "split": "train"}
{"id": "cec8312f4e_8", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "cec8312f4e_9", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. brochures\nB. pen holder\nC. monitor\nD. newspaper", "choices": ["A. brochures", "B. pen holder", "C. monitor", "D. newspaper"], "answer": "C. monitor", "target": "mousepad", "split": "train"}
{"id": "cec8312f4e_10", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also to the side of and below the monitor, about 14 cm away. It is further near the newspaper, about 27 cm away.", "target": "mousepad", "split": "train"}
{"id": "cec8312f4e_11", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 14.2}, {\"object\": \"newspaper\", \"relation\": \"near\", \"distance_cm\": 26.9}]}", "target": "mousepad", "split": "train"}
{"id": "cec8312f4e_12", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "cec8312f4e_13", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. newspaper\nB. brochures\nC. mousepad\nD. plant", "choices": ["A. newspaper", "B. brochures", "C. mousepad", "D. plant"], "answer": "A. newspaper", "target": "pen holder", "split": "train"}
{"id": "cec8312f4e_14", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the table. It is also next to the newspaper, about 14 cm away. It is further below the brochures, about 26 cm away.", "target": "pen holder", "split": "train"}
{"id": "cec8312f4e_15", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"newspaper\", \"relation\": \"next_to\", \"distance_cm\": 13.9}, {\"object\": \"brochures\", \"relation\": \"below\", \"distance_cm\": 26.1}]}", "target": "pen holder", "split": "train"}
{"id": "cec8312f4e_16", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "No", "target": "extension cord", "split": "train"}
{"id": "cec8312f4e_17", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the stapler in this scene?\nA. vase\nB. plant pot\nC. heater\nD. plant", "choices": ["A. vase", "B. plant pot", "C. heater", "D. plant"], "answer": "C. heater", "target": "stapler", "split": "train"}
{"id": "cec8312f4e_18", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is attached to the window. It is also to the side of and above the heater, about 19 cm away. It is further to the side of and below the plant, about 69 cm away.", "target": "stapler", "split": "train"}
{"id": "cec8312f4e_19", "scene_id": "cec8312f4e", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"window\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 18.9}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 69.2}]}", "target": "stapler", "split": "train"}
{"id": "cf1ffd871d_0", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "cf1ffd871d_1", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bottle\nB. storage cabinet\nC. bathroom counter\nD. jacket", "choices": ["A. bottle", "B. storage cabinet", "C. bathroom counter", "D. jacket"], "answer": "B. storage cabinet", "target": "backpack", "split": "train"}
{"id": "cf1ffd871d_2", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the storage cabinet, about 5 cm away. It is also to the side of and below the jacket, about 33 cm away.", "target": "backpack", "split": "train"}
{"id": "cf1ffd871d_3", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"storage cabinet\", \"references\": [{\"object\": \"storage cabinet\", \"relation\": \"side_below\", \"distance_cm\": 5.5}, {\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 32.8}]}", "target": "backpack", "split": "train"}
{"id": "cf1ffd871d_4", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "cf1ffd871d_5", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. light\nB. towel\nC. power socket\nD. bathroom counter", "choices": ["A. light", "B. towel", "C. power socket", "D. bathroom counter"], "answer": "A. light", "target": "bottle", "split": "train"}
{"id": "cf1ffd871d_6", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is below the light, about 31 cm away. It is also near the power socket, about 32 cm away. It is further to the side of and below the towel, about 37 cm away.", "target": "bottle", "split": "train"}
{"id": "cf1ffd871d_7", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"light\", \"relation\": \"below\", \"distance_cm\": 30.7}, {\"object\": \"power socket\", \"relation\": \"near\", \"distance_cm\": 32.0}]}", "target": "bottle", "split": "train"}
{"id": "cf1ffd871d_8", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "Yes", "target": "toothbrush", "split": "train"}
{"id": "cf1ffd871d_9", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the toothbrush in this scene?\nA. light\nB. soap bottle\nC. power socket\nD. bathroom counter", "choices": ["A. light", "B. soap bottle", "C. power socket", "D. bathroom counter"], "answer": "C. power socket", "target": "toothbrush", "split": "train"}
{"id": "cf1ffd871d_10", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the toothbrush in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toothbrush is near the power socket, about 24 cm away. It is also below the light, about 26 cm away. It is further to the side of and above the soap bottle, about 39 cm away.", "target": "toothbrush", "split": "train"}
{"id": "cf1ffd871d_11", "scene_id": "cf1ffd871d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the toothbrush in structured JSON format.", "choices": null, "answer": "{\"target\": \"toothbrush\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"power socket\", \"relation\": \"near\", \"distance_cm\": 23.8}, {\"object\": \"light\", \"relation\": \"below\", \"distance_cm\": 25.8}]}", "target": "toothbrush", "split": "train"}
{"id": "d054227009_0", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "d054227009_1", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. mouse\nB. monitor\nC. keyboard\nD. headphones", "choices": ["A. mouse", "B. monitor", "C. keyboard", "D. headphones"], "answer": "A. mouse", "target": "bottle", "split": "train"}
{"id": "d054227009_2", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the mouse, about 4 cm away. It is further next to the keyboard, about 8 cm away.", "target": "bottle", "split": "train"}
{"id": "d054227009_3", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.6}]}", "target": "bottle", "split": "train"}
{"id": "d054227009_4", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "d054227009_5", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the headphones. It is further next to the laptop.", "target": "keyboard", "split": "train"}
{"id": "d054227009_6", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 2.1}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 2.2}]}", "target": "keyboard", "split": "train"}
{"id": "d054227009_7", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "d054227009_8", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. headphones\nC. monitor\nD. bottle", "choices": ["A. keyboard", "B. headphones", "C. monitor", "D. bottle"], "answer": "D. bottle", "target": "mouse", "split": "train"}
{"id": "d054227009_9", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the bottle, about 4 cm away. It is further next to the keyboard, about 11 cm away.", "target": "mouse", "split": "train"}
{"id": "d054227009_10", "scene_id": "d054227009", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 11.1}]}", "target": "mouse", "split": "train"}
{"id": "d0cb61bc64_0", "scene_id": "d0cb61bc64", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "d0cb61bc64_1", "scene_id": "d0cb61bc64", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. shower wall\nB. shower curtain\nC. shower floor\nD. mop", "choices": ["A. shower wall", "B. shower curtain", "C. shower floor", "D. mop"], "answer": "D. mop", "target": "bottle", "split": "train"}
{"id": "d0cb61bc64_2", "scene_id": "d0cb61bc64", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the mop. It is also to the side of and below the shower curtain, about 12 cm away. It is further to the side of and above the shower floor, about 14 cm away.", "target": "bottle", "split": "train"}
{"id": "d0cb61bc64_3", "scene_id": "d0cb61bc64", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shower floor\", \"references\": [{\"object\": \"mop\", \"relation\": \"under\", \"distance_cm\": 2.2}, {\"object\": \"shower curtain\", \"relation\": \"side_below\", \"distance_cm\": 11.6}]}", "target": "bottle", "split": "train"}
{"id": "d110b8f8b0_0", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "d110b8f8b0_1", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the heat pipe, about 18 cm away. It is also to the side of and below the heater, about 19 cm away. It is further to the side of and below the windowsill, about 47 cm away.", "target": "backpack", "split": "train"}
{"id": "d110b8f8b0_2", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cardboard box\", \"references\": [{\"object\": \"heat pipe\", \"relation\": \"side_above\", \"distance_cm\": 18.2}, {\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 19.0}]}", "target": "backpack", "split": "train"}
{"id": "d110b8f8b0_3", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "d110b8f8b0_4", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. book\nB. electric kettle\nC. thermos\nD. plastic bottle", "choices": ["A. book", "B. electric kettle", "C. thermos", "D. plastic bottle"], "answer": "C. thermos", "target": "bottle", "split": "train"}
{"id": "d110b8f8b0_5", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the window. It is also near the electric kettle, about 16 cm away. It is further to the side of and below the plastic bottle, about 16 cm away.", "target": "bottle", "split": "train"}
{"id": "d110b8f8b0_6", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"clothes drying rack\", \"references\": [{\"object\": \"window\", \"relation\": \"under\", \"distance_cm\": 2.4}, {\"object\": \"electric kettle\", \"relation\": \"near\", \"distance_cm\": 16.1}]}", "target": "bottle", "split": "train"}
{"id": "d110b8f8b0_7", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "d110b8f8b0_8", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the laptop, about 7 cm away. It is further next to the tissue box, about 7 cm away.", "target": "cup", "split": "train"}
{"id": "d110b8f8b0_9", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"tissue box\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "cup", "split": "train"}
{"id": "d110b8f8b0_10", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "train"}
{"id": "d110b8f8b0_11", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. perfume bottle\nB. books\nC. office chair\nD. decoration", "choices": ["A. perfume bottle", "B. books", "C. office chair", "D. decoration"], "answer": "B. books", "target": "extension cord", "split": "train"}
{"id": "d110b8f8b0_12", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the desk. It is also near the books, about 22 cm away. It is further to the side of and above the perfume bottle, about 32 cm away.", "target": "extension cord", "split": "train"}
{"id": "d110b8f8b0_13", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"books\", \"relation\": \"near\", \"distance_cm\": 22.0}, {\"object\": \"perfume bottle\", \"relation\": \"side_above\", \"distance_cm\": 31.8}]}", "target": "extension cord", "split": "train"}
{"id": "d110b8f8b0_14", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "d110b8f8b0_15", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. shelf\nB. desk\nC. office chair\nD. extension cord", "choices": ["A. shelf", "B. desk", "C. office chair", "D. extension cord"], "answer": "B. desk", "target": "laptop charger", "split": "train"}
{"id": "d110b8f8b0_16", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the desk, about 5 cm away. It is also to the side of and below the shelf, about 12 cm away. It is further to the side of and below the office chair, about 41 cm away.", "target": "laptop charger", "split": "train"}
{"id": "d110b8f8b0_17", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 5.3}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 12.2}]}", "target": "laptop charger", "split": "train"}
{"id": "d110b8f8b0_18", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "d110b8f8b0_19", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. stationery\nB. laptop\nC. cup\nD. pens", "choices": ["A. stationery", "B. laptop", "C. cup", "D. pens"], "answer": "A. stationery", "target": "tissue box", "split": "train"}
{"id": "d110b8f8b0_20", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the desk. It is also next to the stationery. It is further next to the cup, about 7 cm away.", "target": "tissue box", "split": "train"}
{"id": "d110b8f8b0_21", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"stationery\", \"relation\": \"next_to\", \"distance_cm\": 2.2}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "tissue box", "split": "train"}
{"id": "d110b8f8b0_22", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "Yes", "target": "tote bag", "split": "train"}
{"id": "d110b8f8b0_23", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. heat pipe\nB. heater\nC. cardboard box\nD. windowsill", "choices": ["A. heat pipe", "B. heater", "C. cardboard box", "D. windowsill"], "answer": "C. cardboard box", "target": "tote bag", "split": "train"}
{"id": "d110b8f8b0_24", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and above the cardboard box, about 9 cm away. It is also to the side of and below the windowsill, about 14 cm away. It is further near the heater, about 18 cm away.", "target": "tote bag", "split": "train"}
{"id": "d110b8f8b0_25", "scene_id": "d110b8f8b0", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"backpack\", \"references\": [{\"object\": \"cardboard box\", \"relation\": \"side_above\", \"distance_cm\": 9.4}, {\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 13.5}]}", "target": "tote bag", "split": "train"}
{"id": "d1f82299d0_0", "scene_id": "d1f82299d0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "d1f82299d0_1", "scene_id": "d1f82299d0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. kitchen counter\nB. kitchen countertop\nC. cupboard\nD. kettle", "choices": ["A. kitchen counter", "B. kitchen countertop", "C. cupboard", "D. kettle"], "answer": "B. kitchen countertop", "target": "cup", "split": "train"}
{"id": "d1f82299d0_2", "scene_id": "d1f82299d0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the kitchen countertop, about 3 cm away. It is also to the side of and above the kitchen counter, about 16 cm away. It is further near the kettle, about 29 cm away.", "target": "cup", "split": "train"}
{"id": "d1f82299d0_3", "scene_id": "d1f82299d0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen countertop\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 16.1}]}", "target": "cup", "split": "train"}
{"id": "d240136ce4_0", "scene_id": "d240136ce4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "d240136ce4_1", "scene_id": "d240136ce4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. power socket\nB. tv\nC. crate\nD. dish soap bottle", "choices": ["A. power socket", "B. tv", "C. crate", "D. dish soap bottle"], "answer": "C. crate", "target": "tv remote", "split": "train"}
{"id": "d240136ce4_2", "scene_id": "d240136ce4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is to the side of and above the crate, about 34 cm away. It is also to the side of and below the tv, about 58 cm away. It is further to the side of and below the power socket, about 59 cm away.", "target": "tv remote", "split": "train"}
{"id": "d240136ce4_3", "scene_id": "d240136ce4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"crate\", \"relation\": \"side_above\", \"distance_cm\": 33.7}, {\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 57.6}]}", "target": "tv remote", "split": "train"}
{"id": "d27235711b_0", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "d27235711b_1", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. books\nB. speaker\nC. vase\nD. bluetooth speaker", "choices": ["A. books", "B. speaker", "C. vase", "D. bluetooth speaker"], "answer": "A. books", "target": "laptop charger", "split": "train"}
{"id": "d27235711b_2", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is on the shelf. It is also to the side of and above the books, about 3 cm away. It is further to the side of and below the vase, about 5 cm away.", "target": "laptop charger", "split": "train"}
{"id": "d27235711b_3", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 3.3}, {\"object\": \"vase\", \"relation\": \"side_below\", \"distance_cm\": 5.5}]}", "target": "laptop charger", "split": "train"}
{"id": "d27235711b_4", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "d27235711b_5", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. banana\nB. pillow\nC. chocolate bar\nD. sofa", "choices": ["A. banana", "B. pillow", "C. chocolate bar", "D. sofa"], "answer": "C. chocolate bar", "target": "mug", "split": "train"}
{"id": "d27235711b_6", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also near the chocolate bar, about 22 cm away. It is further near the banana, about 32 cm away.", "target": "mug", "split": "train"}
{"id": "d27235711b_7", "scene_id": "d27235711b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chocolate bar\", \"relation\": \"near\", \"distance_cm\": 21.7}, {\"object\": \"banana\", \"relation\": \"near\", \"distance_cm\": 32.0}]}", "target": "mug", "split": "train"}
{"id": "d2f44bf242_0", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "d2f44bf242_1", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. table\nB. keyboard\nC. mouse\nD. table lamp", "choices": ["A. table", "B. keyboard", "C. mouse", "D. table lamp"], "answer": "C. mouse", "target": "cable", "split": "train"}
{"id": "d2f44bf242_2", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is on the table. It is also near the mouse, about 21 cm away. It is further near the keyboard, about 24 cm away.", "target": "cable", "split": "train"}
{"id": "d2f44bf242_3", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 21.2}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 24.3}]}", "target": "cable", "split": "train"}
{"id": "d2f44bf242_4", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "d2f44bf242_5", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. table\nB. computer tower\nC. bag\nD. table fan", "choices": ["A. table", "B. computer tower", "C. bag", "D. table fan"], "answer": "D. table fan", "target": "pen", "split": "train"}
{"id": "d2f44bf242_6", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the table. It is also to the side of and below the table fan, about 5 cm away. It is further to the side of and below the bag, about 16 cm away.", "target": "pen", "split": "train"}
{"id": "d2f44bf242_7", "scene_id": "d2f44bf242", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table fan\", \"relation\": \"side_below\", \"distance_cm\": 5.0}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 15.6}]}", "target": "pen", "split": "train"}
{"id": "d3b7d054c4_0", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "d3b7d054c4_1", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. bookshelf\nB. mouse\nC. display cabinet\nD. pencil holder", "choices": ["A. bookshelf", "B. mouse", "C. display cabinet", "D. pencil holder"], "answer": "A. bookshelf", "target": "extension cord", "split": "train"}
{"id": "d3b7d054c4_2", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the desk. It is also under the bookshelf, about 4 cm away. It is further next to the mouse, about 6 cm away.", "target": "extension cord", "split": "train"}
{"id": "d3b7d054c4_3", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"under\", \"distance_cm\": 3.9}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.5}]}", "target": "extension cord", "split": "train"}
{"id": "d3b7d054c4_4", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lanyard appear?", "choices": null, "answer": "Yes", "target": "lanyard", "split": "train"}
{"id": "d3b7d054c4_5", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the lanyard in this scene?\nA. baseball cap\nB. decoration\nC. umbrella\nD. router", "choices": ["A. baseball cap", "B. decoration", "C. umbrella", "D. router"], "answer": "D. router", "target": "lanyard", "split": "train"}
{"id": "d3b7d054c4_6", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the lanyard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The lanyard is to the side of and above the router, about 3 cm away. It is also next to the umbrella, about 11 cm away. It is further below the baseball cap, about 15 cm away.", "target": "lanyard", "split": "train"}
{"id": "d3b7d054c4_7", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the lanyard in structured JSON format.", "choices": null, "answer": "{\"target\": \"lanyard\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"router\", \"relation\": \"side_above\", \"distance_cm\": 2.5}, {\"object\": \"umbrella\", \"relation\": \"next_to\", \"distance_cm\": 11.4}]}", "target": "lanyard", "split": "train"}
{"id": "d3b7d054c4_8", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "No", "target": "key", "split": "train"}
{"id": "d3b7d054c4_9", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the laptop, about 3 cm away. It is also to the side of and above the desk, about 4 cm away. It is further next to the extension cord, about 6 cm away.", "target": "mouse", "split": "train"}
{"id": "d3b7d054c4_10", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 3.9}]}", "target": "mouse", "split": "train"}
{"id": "d3b7d054c4_11", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "d3b7d054c4_12", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the desk. It is also under the monitor, about 3 cm away. It is further under the pencil holder, about 4 cm away.", "target": "mousepad", "split": "train"}
{"id": "d3b7d054c4_13", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.2}, {\"object\": \"pencil holder\", \"relation\": \"under\", \"distance_cm\": 3.8}]}", "target": "mousepad", "split": "train"}
{"id": "d3b7d054c4_14", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "d3b7d054c4_15", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil holder in this scene?\nA. laptop stand\nB. mousepad\nC. extension cord\nD. monitor", "choices": ["A. laptop stand", "B. mousepad", "C. extension cord", "D. monitor"], "answer": "B. mousepad", "target": "pencil holder", "split": "train"}
{"id": "d3b7d054c4_16", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil holder is on the desk. It is also next to the mousepad, about 4 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "pencil holder", "split": "train"}
{"id": "d3b7d054c4_17", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil holder\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 3.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.8}]}", "target": "pencil holder", "split": "train"}
{"id": "d3b7d054c4_18", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "d3b7d054c4_19", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. pillow\nB. pedestal fan\nC. window frame\nD. basket", "choices": ["A. pillow", "B. pedestal fan", "C. window frame", "D. basket"], "answer": "C. window frame", "target": "tissue box", "split": "train"}
{"id": "d3b7d054c4_20", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and below the window frame, about 7 cm away. It is also above the basket, about 11 cm away. It is further to the side of and below the pillow, about 19 cm away.", "target": "tissue box", "split": "train"}
{"id": "d3b7d054c4_21", "scene_id": "d3b7d054c4", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"window frame\", \"relation\": \"side_below\", \"distance_cm\": 7.3}, {\"object\": \"basket\", \"relation\": \"above\", \"distance_cm\": 11.1}]}", "target": "tissue box", "split": "train"}
{"id": "d415cc449b_0", "scene_id": "d415cc449b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "d497bb124d_0", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "d497bb124d_1", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. blanket\nC. book\nD. chair", "choices": ["A. bag", "B. blanket", "C. book", "D. chair"], "answer": "B. blanket", "target": "backpack", "split": "train"}
{"id": "d497bb124d_2", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the blanket, about 44 cm away.", "target": "backpack", "split": "train"}
{"id": "d497bb124d_3", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"plastic bag\", \"references\": [{\"object\": \"blanket\", \"relation\": \"side_below\", \"distance_cm\": 44.2}]}", "target": "backpack", "split": "train"}
{"id": "d497bb124d_4", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "d497bb124d_5", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. sticky notes\nB. tablet\nC. desk\nD. laptop", "choices": ["A. sticky notes", "B. tablet", "C. desk", "D. laptop"], "answer": "C. desk", "target": "mouse", "split": "train"}
{"id": "d497bb124d_6", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the desk, about 4 cm away. It is also to the side of and below the tablet, about 8 cm away. It is further to the side of and below the laptop, about 13 cm away.", "target": "mouse", "split": "train"}
{"id": "d497bb124d_7", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 3.7}, {\"object\": \"tablet\", \"relation\": \"side_below\", \"distance_cm\": 8.3}]}", "target": "mouse", "split": "train"}
{"id": "d497bb124d_8", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "d497bb124d_9", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. sticky notes\nB. laptop\nC. pen\nD. tablet", "choices": ["A. sticky notes", "B. laptop", "C. pen", "D. tablet"], "answer": "D. tablet", "target": "mousepad", "split": "train"}
{"id": "d497bb124d_10", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the desk. It is also under the tablet. It is further to the side of and below the laptop, about 7 cm away.", "target": "mousepad", "split": "train"}
{"id": "d497bb124d_11", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"tablet\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 7.3}]}", "target": "mousepad", "split": "train"}
{"id": "d497bb124d_12", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless headphones appear?", "choices": null, "answer": "No", "target": "wireless headphones", "split": "train"}
{"id": "d497bb124d_13", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. shelf\nB. mousepad\nC. tablet\nD. mouse", "choices": ["A. shelf", "B. mousepad", "C. tablet", "D. mouse"], "answer": "C. tablet", "target": "pen", "split": "train"}
{"id": "d497bb124d_14", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the desk. It is also to the side of and below the tablet, about 9 cm away. It is further next to the mousepad, about 13 cm away.", "target": "pen", "split": "train"}
{"id": "d497bb124d_15", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"tablet\", \"relation\": \"side_below\", \"distance_cm\": 9.0}, {\"object\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 12.6}]}", "target": "pen", "split": "train"}
{"id": "d497bb124d_16", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "Yes", "target": "pen holder", "split": "train"}
{"id": "d497bb124d_17", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. magazines\nB. tablet\nC. desk\nD. book", "choices": ["A. magazines", "B. tablet", "C. desk", "D. book"], "answer": "C. desk", "target": "pen holder", "split": "train"}
{"id": "d497bb124d_18", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is on the shelf. It is also to the side of and above the desk. It is further next to the book, about 6 cm away.", "target": "pen holder", "split": "train"}
{"id": "d497bb124d_19", "scene_id": "d497bb124d", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 2.5}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 6.4}]}", "target": "pen holder", "split": "train"}
{"id": "d537ef1d41_0", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "d537ef1d41_1", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. trash can lid\nB. cardboard box\nC. t-shirt\nD. power socket", "choices": ["A. trash can lid", "B. cardboard box", "C. t-shirt", "D. power socket"], "answer": "D. power socket", "target": "backpack", "split": "train"}
{"id": "d537ef1d41_2", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is near the power socket, about 23 cm away. It is also to the side of and above the trash can lid, about 30 cm away. It is further near the t-shirt, about 39 cm away.", "target": "backpack", "split": "train"}
{"id": "d537ef1d41_3", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cardboard box\", \"references\": [{\"object\": \"power socket\", \"relation\": \"near\", \"distance_cm\": 22.9}, {\"object\": \"trash can lid\", \"relation\": \"side_above\", \"distance_cm\": 29.9}]}", "target": "backpack", "split": "train"}
{"id": "d537ef1d41_4", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "train"}
{"id": "d537ef1d41_5", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. reusable shopping bag\nB. towel\nC. can\nD. dispenser bottle", "choices": ["A. reusable shopping bag", "B. towel", "C. can", "D. dispenser bottle"], "answer": "C. can", "target": "tissue box", "split": "train"}
{"id": "d537ef1d41_6", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the desk. It is also next to the can, about 10 cm away. It is further near the towel, about 18 cm away.", "target": "tissue box", "split": "train"}
{"id": "d537ef1d41_7", "scene_id": "d537ef1d41", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"can\", \"relation\": \"next_to\", \"distance_cm\": 9.8}, {\"object\": \"towel\", \"relation\": \"near\", \"distance_cm\": 18.2}]}", "target": "tissue box", "split": "train"}
{"id": "d593a4f87a_0", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "d593a4f87a_1", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. table organizer\nB. file cabinet\nC. table\nD. printer", "choices": ["A. table organizer", "B. file cabinet", "C. table", "D. printer"], "answer": "C. table", "target": "keyboard", "split": "train"}
{"id": "d593a4f87a_2", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the printer, about 24 cm away. It is also to the side of and above the file cabinet, about 25 cm away. It is further near the mouse, about 28 cm away.", "target": "keyboard", "split": "train"}
{"id": "d593a4f87a_3", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"printer\", \"relation\": \"side_below\", \"distance_cm\": 23.5}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 25.4}]}", "target": "keyboard", "split": "train"}
{"id": "d593a4f87a_4", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "d593a4f87a_5", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. table organizer\nC. file cabinet\nD. keyboard", "choices": ["A. office chair", "B. table organizer", "C. file cabinet", "D. keyboard"], "answer": "A. office chair", "target": "mouse", "split": "train"}
{"id": "d593a4f87a_6", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the office chair, about 12 cm away. It is also to the side of and below the table organizer, about 15 cm away. It is further above the file cabinet, about 16 cm away.", "target": "mouse", "split": "train"}
{"id": "d593a4f87a_7", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.3}, {\"object\": \"table organizer\", \"relation\": \"side_below\", \"distance_cm\": 15.4}]}", "target": "mouse", "split": "train"}
{"id": "d593a4f87a_8", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "d593a4f87a_9", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. office chair\nB. file folder\nC. cabinet\nD. file cabinet", "choices": ["A. office chair", "B. file folder", "C. cabinet", "D. file cabinet"], "answer": "A. office chair", "target": "mouse pad", "split": "train"}
{"id": "d593a4f87a_10", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is to the side of and above the office chair, about 5 cm away. It is also next to the file folder, about 13 cm away. It is further to the side of and above the file cabinet, about 22 cm away.", "target": "mouse pad", "split": "train"}
{"id": "d593a4f87a_11", "scene_id": "d593a4f87a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 4.9}, {\"object\": \"file folder\", \"relation\": \"next_to\", \"distance_cm\": 13.4}]}", "target": "mouse pad", "split": "train"}
{"id": "d61691f945_0", "scene_id": "d61691f945", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "d6a77f7c22_0", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "d6a77f7c22_1", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. sofa\nB. coaster\nC. kitchen paper\nD. cushion", "choices": ["A. sofa", "B. coaster", "C. kitchen paper", "D. cushion"], "answer": "C. kitchen paper", "target": "mug", "split": "train"}
{"id": "d6a77f7c22_2", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the coffee table. It is also under the kitchen paper, about 4 cm away. It is further near the coaster, about 25 cm away.", "target": "mug", "split": "train"}
{"id": "d6a77f7c22_3", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"kitchen paper\", \"relation\": \"under\", \"distance_cm\": 4.2}, {\"object\": \"coaster\", \"relation\": \"near\", \"distance_cm\": 24.7}]}", "target": "mug", "split": "train"}
{"id": "d6a77f7c22_4", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "d6a77f7c22_5", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. refrigerator\nB. dish rack\nC. cutting board\nD. power socket", "choices": ["A. refrigerator", "B. dish rack", "C. cutting board", "D. power socket"], "answer": "A. refrigerator", "target": "power strip", "split": "train"}
{"id": "d6a77f7c22_6", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and above the refrigerator, about 9 cm away. It is also near the dish rack, about 16 cm away. It is further to the side of and below the power socket, about 19 cm away.", "target": "power strip", "split": "train"}
{"id": "d6a77f7c22_7", "scene_id": "d6a77f7c22", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"refrigerator\", \"relation\": \"side_above\", \"distance_cm\": 8.6}, {\"object\": \"dish rack\", \"relation\": \"near\", \"distance_cm\": 16.5}]}", "target": "power strip", "split": "train"}
{"id": "d6bb698875_0", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "d6bb698875_1", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. jacket\nB. doorframe\nC. poster\nD. sling bag", "choices": ["A. jacket", "B. doorframe", "C. poster", "D. sling bag"], "answer": "B. doorframe", "target": "backpack", "split": "train"}
{"id": "d6bb698875_2", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the doorframe, about 11 cm away. It is also to the side of and below the door, about 13 cm away. It is further to the side of and below the jacket, about 15 cm away.", "target": "backpack", "split": "train"}
{"id": "d6bb698875_3", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"door\", \"references\": [{\"object\": \"doorframe\", \"relation\": \"side_below\", \"distance_cm\": 11.2}, {\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 12.5}]}", "target": "backpack", "split": "train"}
{"id": "d6bb698875_4", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "Yes", "target": "glasses case", "split": "train"}
{"id": "d6bb698875_5", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. books\nB. yoga mat\nC. bookshelf\nD. board", "choices": ["A. books", "B. yoga mat", "C. bookshelf", "D. board"], "answer": "C. bookshelf", "target": "glasses case", "split": "train"}
{"id": "d6bb698875_6", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is on the cabinet. It is also to the side of and below the bookshelf, about 6 cm away. It is further to the side of and above the books, about 14 cm away.", "target": "glasses case", "split": "train"}
{"id": "d6bb698875_7", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 6.2}, {\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 13.8}]}", "target": "glasses case", "split": "train"}
{"id": "d6bb698875_8", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "d6bb698875_9", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the table, about 10 cm away. It is also next to the desk lamp, about 10 cm away. It is further to the side of and above the bottle, about 15 cm away.", "target": "headphones", "split": "train"}
{"id": "d6bb698875_10", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk lamp\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 10.1}, {\"object\": \"desk lamp\", \"relation\": \"next_to\", \"distance_cm\": 10.2}]}", "target": "headphones", "split": "train"}
{"id": "d6bb698875_11", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the key appear?", "choices": null, "answer": "Yes", "target": "key", "split": "train"}
{"id": "d6bb698875_12", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the key in this scene?\nA. yoga mat\nB. wallet\nC. books\nD. bookshelf", "choices": ["A. yoga mat", "B. wallet", "C. books", "D. bookshelf"], "answer": "B. wallet", "target": "key", "split": "train"}
{"id": "d6bb698875_13", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the key in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The key is on the cabinet. It is also to the side of and above the yoga mat, about 15 cm away. It is further to the side of and below the bookshelf, about 20 cm away.", "target": "key", "split": "train"}
{"id": "d6bb698875_14", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the key in structured JSON format.", "choices": null, "answer": "{\"target\": \"key\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"yoga mat\", \"relation\": \"side_above\", \"distance_cm\": 14.6}, {\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 19.7}]}", "target": "key", "split": "train"}
{"id": "d6bb698875_15", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "d6bb698875_16", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. laptop stand\nB. monitor\nC. external memory\nD. bottle", "choices": ["A. laptop stand", "B. monitor", "C. external memory", "D. bottle"], "answer": "C. external memory", "target": "keyboard", "split": "train"}
{"id": "d6bb698875_17", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the external memory, about 4 cm away. It is further to the side of and below the monitor, about 8 cm away.", "target": "keyboard", "split": "train"}
{"id": "d6bb698875_18", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"external memory\", \"relation\": \"next_to\", \"distance_cm\": 3.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.6}]}", "target": "keyboard", "split": "train"}
{"id": "d6bb698875_19", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "d6bb698875_20", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. stationery\nB. bottle\nC. laptop stand\nD. pensil case", "choices": ["A. stationery", "B. bottle", "C. laptop stand", "D. pensil case"], "answer": "A. stationery", "target": "mouse", "split": "train"}
{"id": "d6bb698875_21", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and below the stationery, about 9 cm away. It is further next to the laptop stand, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "d6bb698875_22", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"stationery\", \"relation\": \"side_below\", \"distance_cm\": 8.8}, {\"object\": \"laptop stand\", \"relation\": \"next_to\", \"distance_cm\": 13.7}]}", "target": "mouse", "split": "train"}
{"id": "d6bb698875_23", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "d6bb698875_24", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. backpack\nB. decoration\nC. book\nD. box", "choices": ["A. backpack", "B. decoration", "C. book", "D. box"], "answer": "D. box", "target": "mug", "split": "train"}
{"id": "d6bb698875_25", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the coffee table. It is also next to the box, about 5 cm away. It is further near the decoration, about 16 cm away.", "target": "mug", "split": "train"}
{"id": "d6bb698875_26", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"coffee table\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 4.8}, {\"object\": \"decoration\", \"relation\": \"near\", \"distance_cm\": 16.4}]}", "target": "mug", "split": "train"}
{"id": "d6bb698875_27", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "d6bb698875_28", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. pillow\nB. plant\nC. hiking poles\nD. windowsill", "choices": ["A. pillow", "B. plant", "C. hiking poles", "D. windowsill"], "answer": "B. plant", "target": "power strip", "split": "train"}
{"id": "d6bb698875_29", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the plant. It is also to the side of and above the hiking poles, about 35 cm away. It is further to the side of and above the pillow, about 35 cm away.", "target": "power strip", "split": "train"}
{"id": "d6bb698875_30", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"side table\", \"references\": [{\"object\": \"plant\", \"relation\": \"under\", \"distance_cm\": 2.4}, {\"object\": \"hiking poles\", \"relation\": \"side_above\", \"distance_cm\": 35.3}]}", "target": "power strip", "split": "train"}
{"id": "d6bb698875_31", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tape dispenser appear?", "choices": null, "answer": "No", "target": "tape dispenser", "split": "train"}
{"id": "d6bb698875_32", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue in this scene?\nA. suitcase\nB. trash can\nC. orginizer\nD. picture", "choices": ["A. suitcase", "B. trash can", "C. orginizer", "D. picture"], "answer": "C. orginizer", "target": "tissue", "split": "train"}
{"id": "d6bb698875_33", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "33", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue is to the side of and above the orginizer, about 26 cm away. It is also to the side of and below the window, about 27 cm away. It is further to the side of and above the trash can, about 36 cm away.", "target": "tissue", "split": "train"}
{"id": "d6bb698875_34", "scene_id": "d6bb698875", "source_dataset": "scannetpp", "local_sample_id": "34", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"orginizer\", \"relation\": \"side_above\", \"distance_cm\": 25.6}, {\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 27.2}]}", "target": "tissue", "split": "train"}
{"id": "d7abfc4b17_0", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "d7abfc4b17_1", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. plant\nB. bathroom cabinet\nC. tap\nD. sink", "choices": ["A. plant", "B. bathroom cabinet", "C. tap", "D. sink"], "answer": "D. sink", "target": "bottle", "split": "train"}
{"id": "d7abfc4b17_2", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the sink, about 4 cm away. It is also to the side of and above the tap, about 16 cm away. It is further to the side of and above the bathroom cabinet, about 37 cm away.", "target": "bottle", "split": "train"}
{"id": "d7abfc4b17_3", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 3.8}, {\"object\": \"tap\", \"relation\": \"side_above\", \"distance_cm\": 16.3}]}", "target": "bottle", "split": "train"}
{"id": "d7abfc4b17_4", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "No", "target": "keyboard", "split": "train"}
{"id": "d7abfc4b17_5", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. plant pot\nB. rack\nC. toilet paper\nD. plant", "choices": ["A. plant pot", "B. rack", "C. toilet paper", "D. plant"], "answer": "C. toilet paper", "target": "tissue box", "split": "train"}
{"id": "d7abfc4b17_6", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is to the side of and above the toilet paper, about 10 cm away. It is also to the side of and below the rack, about 21 cm away. It is further to the side of and above the plant, about 26 cm away.", "target": "tissue box", "split": "train"}
{"id": "d7abfc4b17_7", "scene_id": "d7abfc4b17", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"rack\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"side_above\", \"distance_cm\": 9.9}, {\"object\": \"rack\", \"relation\": \"side_below\", \"distance_cm\": 20.8}]}", "target": "tissue box", "split": "train"}
{"id": "d7b871aaa8_0", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "d7b871aaa8_1", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. telephone\nB. power socket\nC. monitor\nD. windowsill", "choices": ["A. telephone", "B. power socket", "C. monitor", "D. windowsill"], "answer": "C. monitor", "target": "keyboard", "split": "train"}
{"id": "d7b871aaa8_2", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also under the monitor, about 3 cm away. It is further to the side of and below the power socket, about 11 cm away.", "target": "keyboard", "split": "train"}
{"id": "d7b871aaa8_3", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.4}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 10.6}]}", "target": "keyboard", "split": "train"}
{"id": "d7b871aaa8_4", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "d7b871aaa8_5", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. keyboard\nC. table\nD. monitor", "choices": ["A. chair", "B. keyboard", "C. table", "D. monitor"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "d7b871aaa8_6", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 8 cm away. It is further to the side of and below the monitor, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "d7b871aaa8_7", "scene_id": "d7b871aaa8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.5}]}", "target": "mouse", "split": "train"}
{"id": "d807fb583b_0", "scene_id": "d807fb583b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "d807fb583b_1", "scene_id": "d807fb583b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. computer tower\nB. table\nC. chair\nD. book", "choices": ["A. computer tower", "B. table", "C. chair", "D. book"], "answer": "B. table", "target": "power adapter", "split": "train"}
{"id": "d807fb583b_2", "scene_id": "d807fb583b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is on the table. It is also to the side of and below the computer tower, about 43 cm away. It is further to the side of and above the chair, about 49 cm away.", "target": "power adapter", "split": "train"}
{"id": "d807fb583b_3", "scene_id": "d807fb583b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_below\", \"distance_cm\": 43.1}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 48.8}]}", "target": "power adapter", "split": "train"}
{"id": "d902931864_0", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "d902931864_1", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. hanger rod\nB. clothes\nC. jacket\nD. door frame", "choices": ["A. hanger rod", "B. clothes", "C. jacket", "D. door frame"], "answer": "A. hanger rod", "target": "backpack", "split": "train"}
{"id": "d902931864_2", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is under the hanger rod. It is also to the side of and above the door frame, about 25 cm away. It is further to the side of and above the door, about 28 cm away.", "target": "backpack", "split": "train"}
{"id": "d902931864_3", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"door\", \"references\": [{\"object\": \"hanger rod\", \"relation\": \"under\", \"distance_cm\": 2.4}, {\"object\": \"door frame\", \"relation\": \"side_above\", \"distance_cm\": 25.0}]}", "target": "backpack", "split": "train"}
{"id": "d902931864_4", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "d902931864_5", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. box\nB. chair\nC. jar\nD. laptop", "choices": ["A. box", "B. chair", "C. jar", "D. laptop"], "answer": "D. laptop", "target": "bottle", "split": "train"}
{"id": "d902931864_6", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also near the laptop, about 26 cm away. It is further to the side of and above the chair, about 28 cm away.", "target": "bottle", "split": "train"}
{"id": "d902931864_7", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 25.8}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 27.6}]}", "target": "bottle", "split": "train"}
{"id": "d902931864_8", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "d902931864_9", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. cabinet\nB. laptop\nC. roller blinds\nD. chocolate bar", "choices": ["A. cabinet", "B. laptop", "C. roller blinds", "D. chocolate bar"], "answer": "B. laptop", "target": "mouse", "split": "train"}
{"id": "d902931864_10", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and below the laptop, about 8 cm away. It is further to the side of and above the cabinet, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "d902931864_11", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 7.8}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 14.0}]}", "target": "mouse", "split": "train"}
{"id": "d902931864_12", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "Yes", "target": "pencil case", "split": "train"}
{"id": "d902931864_13", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil case in this scene?\nA. laptop\nB. jar\nC. chair\nD. box", "choices": ["A. laptop", "B. jar", "C. chair", "D. box"], "answer": "A. laptop", "target": "pencil case", "split": "train"}
{"id": "d902931864_14", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil case is on the table. It is also near the laptop, about 17 cm away. It is further near the jar, about 18 cm away.", "target": "pencil case", "split": "train"}
{"id": "d902931864_15", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 16.7}, {\"object\": \"jar\", \"relation\": \"near\", \"distance_cm\": 18.2}]}", "target": "pencil case", "split": "train"}
{"id": "d902931864_16", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "Yes", "target": "tote bag", "split": "train"}
{"id": "d902931864_17", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. box\nB. bag\nC. light switch\nD. hanger rod", "choices": ["A. box", "B. bag", "C. light switch", "D. hanger rod"], "answer": "A. box", "target": "tote bag", "split": "train"}
{"id": "d902931864_18", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is to the side of and below the box, about 36 cm away. It is also to the side of and below the bag, about 37 cm away. It is further to the side of and below the hanger rod, about 39 cm away.", "target": "tote bag", "split": "train"}
{"id": "d902931864_19", "scene_id": "d902931864", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 35.5}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 37.2}]}", "target": "tote bag", "split": "train"}
{"id": "d986399f4c_0", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "No", "target": "headphones", "split": "train"}
{"id": "d986399f4c_1", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. apples\nB. cup\nC. printing paper\nD. bag", "choices": ["A. apples", "B. cup", "C. printing paper", "D. bag"], "answer": "C. printing paper", "target": "backpack", "split": "train"}
{"id": "d986399f4c_2", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the printing paper, about 19 cm away. It is also to the side of and below the cup, about 55 cm away. It is further near the bag, about 70 cm away.", "target": "backpack", "split": "train"}
{"id": "d986399f4c_3", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"printing paper\", \"references\": [{\"object\": \"printing paper\", \"relation\": \"side_below\", \"distance_cm\": 19.1}, {\"object\": \"cup\", \"relation\": \"side_below\", \"distance_cm\": 55.4}]}", "target": "backpack", "split": "train"}
{"id": "d986399f4c_4", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "d986399f4c_5", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. laptop\nB. mouse pad\nC. laptop stand\nD. mouse", "choices": ["A. laptop", "B. mouse pad", "C. laptop stand", "D. mouse"], "answer": "B. mouse pad", "target": "keyboard", "split": "train"}
{"id": "d986399f4c_6", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse pad, about 7 cm away. It is also next to the laptop, about 12 cm away. It is further next to the mouse, about 12 cm away.", "target": "keyboard", "split": "train"}
{"id": "d986399f4c_7", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse pad\", \"relation\": \"next_to\", \"distance_cm\": 7.5}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 11.5}]}", "target": "keyboard", "split": "train"}
{"id": "d986399f4c_8", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "d986399f4c_9", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. cup\nC. decoration\nD. desk", "choices": ["A. keyboard", "B. cup", "C. decoration", "D. desk"], "answer": "D. desk", "target": "mouse", "split": "train"}
{"id": "d986399f4c_10", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the desk, about 3 cm away. It is also next to the cup, about 10 cm away. It is further next to the keyboard, about 12 cm away.", "target": "mouse", "split": "train"}
{"id": "d986399f4c_11", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 3.1}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 9.5}]}", "target": "mouse", "split": "train"}
{"id": "d986399f4c_12", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "d986399f4c_13", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. cup\nB. electric kettle\nC. keyboard\nD. decoration", "choices": ["A. cup", "B. electric kettle", "C. keyboard", "D. decoration"], "answer": "A. cup", "target": "mouse pad", "split": "train"}
{"id": "d986399f4c_14", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is next to the cup, about 4 cm away. It is also next to the keyboard, about 7 cm away. It is further to the side of and below the decoration, about 11 cm away.", "target": "mouse pad", "split": "train"}
{"id": "d986399f4c_15", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 3.8}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.5}]}", "target": "mouse pad", "split": "train"}
{"id": "d986399f4c_16", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "d986399f4c_17", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. monitor\nB. desk\nC. laptop stand\nD. mouse pad", "choices": ["A. monitor", "B. desk", "C. laptop stand", "D. mouse pad"], "answer": "B. desk", "target": "phone charger", "split": "train"}
{"id": "d986399f4c_18", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and above the desk, about 4 cm away. It is also to the side of and below the monitor, about 14 cm away. It is further near the laptop stand, about 16 cm away.", "target": "phone charger", "split": "train"}
{"id": "d986399f4c_19", "scene_id": "d986399f4c", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 3.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.8}]}", "target": "phone charger", "split": "train"}
{"id": "d9c781301e_0", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "d9c781301e_1", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. cabinet\nB. air purifier\nC. cable\nD. light switch", "choices": ["A. cabinet", "B. air purifier", "C. cable", "D. light switch"], "answer": "A. cabinet", "target": "bottle", "split": "train"}
{"id": "d9c781301e_2", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the cabinet, about 3 cm away. It is also to the side of and above the light switch, about 72 cm away.", "target": "bottle", "split": "train"}
{"id": "d9c781301e_3", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"light switch\", \"relation\": \"side_above\", \"distance_cm\": 72.0}]}", "target": "bottle", "split": "train"}
{"id": "d9c781301e_4", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "d9c781301e_5", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. heater\nB. laptop\nC. power socket\nD. envelope", "choices": ["A. heater", "B. laptop", "C. power socket", "D. envelope"], "answer": "B. laptop", "target": "cable", "split": "train"}
{"id": "d9c781301e_6", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and above the laptop, about 11 cm away. It is also to the side of and above the heater, about 16 cm away. It is further to the side of and below the envelope, about 19 cm away.", "target": "cable", "split": "train"}
{"id": "d9c781301e_7", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"paper\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 11.0}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 15.7}]}", "target": "cable", "split": "train"}
{"id": "d9c781301e_8", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "Yes", "target": "highlighter", "split": "train"}
{"id": "d9c781301e_9", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the highlighter in this scene?\nA. paper\nB. keyboard\nC. hole puncher\nD. cabinet", "choices": ["A. paper", "B. keyboard", "C. hole puncher", "D. cabinet"], "answer": "A. paper", "target": "highlighter", "split": "train"}
{"id": "d9c781301e_10", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the highlighter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The highlighter is on the desk. It is also near the hole puncher, about 24 cm away. It is further to the side of and above the cabinet, about 27 cm away.", "target": "highlighter", "split": "train"}
{"id": "d9c781301e_11", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the highlighter in structured JSON format.", "choices": null, "answer": "{\"target\": \"highlighter\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"hole puncher\", \"relation\": \"near\", \"distance_cm\": 24.5}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 26.9}]}", "target": "highlighter", "split": "train"}
{"id": "d9c781301e_12", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "d9c781301e_13", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. office chair\nC. keyboard\nD. power strip", "choices": ["A. monitor", "B. office chair", "C. keyboard", "D. power strip"], "answer": "C. keyboard", "target": "mouse", "split": "train"}
{"id": "d9c781301e_14", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard. It is also to the side of and below the monitor, about 28 cm away. It is further to the side of and below the power strip, about 31 cm away.", "target": "mouse", "split": "train"}
{"id": "d9c781301e_15", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 27.8}]}", "target": "mouse", "split": "train"}
{"id": "d9c781301e_16", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "d9c781301e_17", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. envelope\nB. power socket\nC. air purifier\nD. heater", "choices": ["A. envelope", "B. power socket", "C. air purifier", "D. heater"], "answer": "D. heater", "target": "notebook", "split": "train"}
{"id": "d9c781301e_18", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and above the heater, about 39 cm away. It is also near the envelope, about 64 cm away. It is further to the side of and above the power socket, about 75 cm away.", "target": "notebook", "split": "train"}
{"id": "d9c781301e_19", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"heater\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 39.3}, {\"object\": \"envelope\", \"relation\": \"near\", \"distance_cm\": 64.3}]}", "target": "notebook", "split": "train"}
{"id": "d9c781301e_20", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "d9c781301e_21", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor\nB. cabinet\nC. monitor arm\nD. file holder", "choices": ["A. monitor", "B. cabinet", "C. monitor arm", "D. file holder"], "answer": "D. file holder", "target": "power strip", "split": "train"}
{"id": "d9c781301e_22", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the desk. It is also to the side of and below the file holder, about 18 cm away. It is further to the side of and below the monitor, about 24 cm away.", "target": "power strip", "split": "train"}
{"id": "d9c781301e_23", "scene_id": "d9c781301e", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"file holder\", \"relation\": \"side_below\", \"distance_cm\": 18.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.5}]}", "target": "power strip", "split": "train"}
{"id": "da1df716a5_0", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "da1df716a5_1", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. power sockets\nB. extension cord\nC. router\nD. phone charger", "choices": ["A. power sockets", "B. extension cord", "C. router", "D. phone charger"], "answer": "C. router", "target": "laptop charger", "split": "train"}
{"id": "da1df716a5_2", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is next to the router, about 5 cm away. It is also to the side of and below the power sockets, about 12 cm away. It is further to the side of and below the phone charger, about 17 cm away.", "target": "laptop charger", "split": "train"}
{"id": "da1df716a5_3", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"router\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 12.4}]}", "target": "laptop charger", "split": "train"}
{"id": "da1df716a5_4", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "da1df716a5_5", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. stationeries\nB. table lamp\nC. notebook\nD. laptop", "choices": ["A. stationeries", "B. table lamp", "C. notebook", "D. laptop"], "answer": "D. laptop", "target": "mouse", "split": "train"}
{"id": "da1df716a5_6", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the laptop, about 12 cm away. It is further next to the stationeries, about 13 cm away.", "target": "mouse", "split": "train"}
{"id": "da1df716a5_7", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 11.5}, {\"object\": \"stationeries\", \"relation\": \"next_to\", \"distance_cm\": 13.1}]}", "target": "mouse", "split": "train"}
{"id": "da1df716a5_8", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "Yes", "target": "post it note", "split": "train"}
{"id": "da1df716a5_9", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the post it note in this scene?\nA. tissue pack\nB. paper\nC. table lamp\nD. stationeries", "choices": ["A. tissue pack", "B. paper", "C. table lamp", "D. stationeries"], "answer": "B. paper", "target": "post it note", "split": "train"}
{"id": "da1df716a5_10", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the post it note in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The post it note is under the paper, about 4 cm away. It is also to the side of and above the table lamp, about 34 cm away. It is further to the side of and above the stationeries, about 35 cm away.", "target": "post it note", "split": "train"}
{"id": "da1df716a5_11", "scene_id": "da1df716a5", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the post it note in structured JSON format.", "choices": null, "answer": "{\"target\": \"post it note\", \"support_surface\": \"table lamp\", \"references\": [{\"object\": \"paper\", \"relation\": \"under\", \"distance_cm\": 3.9}, {\"object\": \"table lamp\", \"relation\": \"side_above\", \"distance_cm\": 33.7}]}", "target": "post it note", "split": "train"}
{"id": "daffc70503_0", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the game controller appear?", "choices": null, "answer": "No", "target": "game controller", "split": "train"}
{"id": "daffc70503_1", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. whiteboard eraser\nB. marker\nC. paper\nD. chair", "choices": ["A. whiteboard eraser", "B. marker", "C. paper", "D. chair"], "answer": "C. paper", "target": "bottle", "split": "train"}
{"id": "daffc70503_2", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the paper, about 10 cm away. It is further to the side of and above the chair, about 26 cm away.", "target": "bottle", "split": "train"}
{"id": "daffc70503_3", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 9.5}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 26.1}]}", "target": "bottle", "split": "train"}
{"id": "daffc70503_4", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "daffc70503_5", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the marker in this scene?\nA. chair\nB. whiteboard eraser\nC. bottle\nD. paper", "choices": ["A. chair", "B. whiteboard eraser", "C. bottle", "D. paper"], "answer": "B. whiteboard eraser", "target": "marker", "split": "train"}
{"id": "daffc70503_6", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the marker in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The marker is on the table. It is also next to the whiteboard eraser, about 10 cm away. It is further next to the paper, about 12 cm away.", "target": "marker", "split": "train"}
{"id": "daffc70503_7", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the marker in structured JSON format.", "choices": null, "answer": "{\"target\": \"marker\", \"support_surface\": \"table\", \"references\": [{\"object\": \"whiteboard eraser\", \"relation\": \"next_to\", \"distance_cm\": 9.6}, {\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 11.9}]}", "target": "marker", "split": "train"}
{"id": "daffc70503_8", "scene_id": "daffc70503", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "db32da9b42_0", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "db32da9b42_1", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. book\nB. books\nC. tower fan\nD. blanket", "choices": ["A. book", "B. books", "C. tower fan", "D. blanket"], "answer": "C. tower fan", "target": "backpack", "split": "train"}
{"id": "db32da9b42_2", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the cabinet. It is also to the side of and above the tower fan, about 23 cm away.", "target": "backpack", "split": "train"}
{"id": "db32da9b42_3", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"tower fan\", \"relation\": \"side_above\", \"distance_cm\": 22.8}]}", "target": "backpack", "split": "train"}
{"id": "db32da9b42_4", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "db32da9b42_5", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. keyboard\nB. chair\nC. monitor\nD. standing lamp", "choices": ["A. keyboard", "B. chair", "C. monitor", "D. standing lamp"], "answer": "A. keyboard", "target": "headphones", "split": "train"}
{"id": "db32da9b42_6", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also next to the keyboard, about 7 cm away. It is further to the side of and above the chair, about 10 cm away.", "target": "headphones", "split": "train"}
{"id": "db32da9b42_7", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 10.2}]}", "target": "headphones", "split": "train"}
{"id": "db32da9b42_8", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "db32da9b42_9", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. table\nB. chair\nC. curtain\nD. monitor", "choices": ["A. table", "B. chair", "C. curtain", "D. monitor"], "answer": "A. table", "target": "mouse", "split": "train"}
{"id": "db32da9b42_10", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the table, about 9 cm away. It is also to the side of and above the chair, about 22 cm away. It is further to the side of and below the monitor, about 22 cm away.", "target": "mouse", "split": "train"}
{"id": "db32da9b42_11", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 9.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 21.7}]}", "target": "mouse", "split": "train"}
{"id": "db32da9b42_12", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "No", "target": "handbag", "split": "train"}
{"id": "db32da9b42_13", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. chair\nB. curtain\nC. keyboard\nD. window sill", "choices": ["A. chair", "B. curtain", "C. keyboard", "D. window sill"], "answer": "C. keyboard", "target": "mousepad", "split": "train"}
{"id": "db32da9b42_14", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also next to the keyboard, about 3 cm away. It is further to the side of and below the curtain, about 6 cm away.", "target": "mousepad", "split": "train"}
{"id": "db32da9b42_15", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.8}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 6.4}]}", "target": "mousepad", "split": "train"}
{"id": "db32da9b42_16", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "No", "target": "notebook", "split": "train"}
{"id": "db32da9b42_17", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power extension in this scene?\nA. pillow\nB. blanket\nC. backpack\nD. book", "choices": ["A. pillow", "B. blanket", "C. backpack", "D. book"], "answer": "B. blanket", "target": "power extension", "split": "train"}
{"id": "db32da9b42_18", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power extension in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power extension is near the blanket, about 32 cm away. It is also to the side of and below the pillow, about 83 cm away.", "target": "power extension", "split": "train"}
{"id": "db32da9b42_19", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power extension in structured JSON format.", "choices": null, "answer": "{\"target\": \"power extension\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"blanket\", \"relation\": \"near\", \"distance_cm\": 31.8}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 82.7}]}", "target": "power extension", "split": "train"}
{"id": "db32da9b42_20", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "db32da9b42_21", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. table\nB. desk lamp\nC. coffee table\nD. plant", "choices": ["A. table", "B. desk lamp", "C. coffee table", "D. plant"], "answer": "A. table", "target": "power strip", "split": "train"}
{"id": "db32da9b42_22", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the table. It is also to the side of and below the plant, about 7 cm away. It is further to the side of and below the coffee table, about 54 cm away.", "target": "power strip", "split": "train"}
{"id": "db32da9b42_23", "scene_id": "db32da9b42", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"under\", \"distance_cm\": 2.4}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 7.3}]}", "target": "power strip", "split": "train"}
{"id": "dc204f59ae_0", "scene_id": "dc204f59ae", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "dc204f59ae_1", "scene_id": "dc204f59ae", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. monitor\nB. paper\nC. barstool\nD. table", "choices": ["A. monitor", "B. paper", "C. barstool", "D. table"], "answer": "B. paper", "target": "cup", "split": "train"}
{"id": "dc204f59ae_2", "scene_id": "dc204f59ae", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and above the paper, about 26 cm away. It is further to the side of and above the barstool, about 34 cm away.", "target": "cup", "split": "train"}
{"id": "dc204f59ae_3", "scene_id": "dc204f59ae", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 25.6}, {\"object\": \"barstool\", \"relation\": \"side_above\", \"distance_cm\": 34.3}]}", "target": "cup", "split": "train"}
{"id": "dd685be466_0", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "dd685be466_1", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bedside table\nB. pillow\nC. bottle\nD. bowl", "choices": ["A. bedside table", "B. pillow", "C. bottle", "D. bowl"], "answer": "B. pillow", "target": "backpack", "split": "train"}
{"id": "dd685be466_2", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the pillow, about 43 cm away.", "target": "backpack", "split": "train"}
{"id": "dd685be466_3", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 42.6}]}", "target": "backpack", "split": "train"}
{"id": "dd685be466_4", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "dd685be466_5", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. shelf\nB. bowl\nC. bedside table\nD. backpack", "choices": ["A. shelf", "B. bowl", "C. bedside table", "D. backpack"], "answer": "A. shelf", "target": "bottle", "split": "train"}
{"id": "dd685be466_6", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the shelf, about 11 cm away.", "target": "bottle", "split": "train"}
{"id": "dd685be466_7", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 10.8}]}", "target": "bottle", "split": "train"}
{"id": "dd685be466_8", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "dd685be466_9", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. curtain\nB. chair\nC. bowl\nD. laptop charger", "choices": ["A. curtain", "B. chair", "C. bowl", "D. laptop charger"], "answer": "B. chair", "target": "cup", "split": "train"}
{"id": "dd685be466_10", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and above the chair, about 3 cm away. It is further next to the bowl, about 10 cm away.", "target": "cup", "split": "train"}
{"id": "dd685be466_11", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 2.7}, {\"object\": \"bowl\", \"relation\": \"next_to\", \"distance_cm\": 9.9}]}", "target": "cup", "split": "train"}
{"id": "dd685be466_12", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "dd685be466_13", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. laptop\nB. cup\nC. pillow\nD. chair", "choices": ["A. laptop", "B. cup", "C. pillow", "D. chair"], "answer": "D. chair", "target": "laptop charger", "split": "train"}
{"id": "dd685be466_14", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is on the table. It is also to the side of and above the chair, about 6 cm away. It is further next to the laptop, about 14 cm away.", "target": "laptop charger", "split": "train"}
{"id": "dd685be466_15", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 6.4}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 14.3}]}", "target": "laptop charger", "split": "train"}
{"id": "dd685be466_16", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "dd685be466_17", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. box\nB. router\nC. coax outlet\nD. standing lamp", "choices": ["A. box", "B. router", "C. coax outlet", "D. standing lamp"], "answer": "C. coax outlet", "target": "phone charger", "split": "train"}
{"id": "dd685be466_18", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is next to the coax outlet, about 12 cm away. It is also to the side of and below the standing lamp, about 57 cm away. It is further to the side of and above the router, about 77 cm away.", "target": "phone charger", "split": "train"}
{"id": "dd685be466_19", "scene_id": "dd685be466", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"box\", \"references\": [{\"object\": \"coax outlet\", \"relation\": \"next_to\", \"distance_cm\": 12.3}, {\"object\": \"standing lamp\", \"relation\": \"side_below\", \"distance_cm\": 57.1}]}", "target": "phone charger", "split": "train"}
{"id": "de3c77cecd_0", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "de3c77cecd_1", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. tissue box\nB. split\nC. headphones\nD. mouse", "choices": ["A. tissue box", "B. split", "C. headphones", "D. mouse"], "answer": "D. mouse", "target": "backpack", "split": "train"}
{"id": "de3c77cecd_2", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is near the mouse, about 45 cm away. It is also near the headphones, about 47 cm away. It is further to the side of and above the split, about 48 cm away.", "target": "backpack", "split": "train"}
{"id": "de3c77cecd_3", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 44.9}, {\"object\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 46.6}]}", "target": "backpack", "split": "train"}
{"id": "de3c77cecd_4", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "de3c77cecd_5", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. paper\nB. mouse\nC. keyboard\nD. box", "choices": ["A. paper", "B. mouse", "C. keyboard", "D. box"], "answer": "C. keyboard", "target": "bottle", "split": "train"}
{"id": "de3c77cecd_6", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the keyboard, about 10 cm away. It is further next to the mouse, about 14 cm away.", "target": "bottle", "split": "train"}
{"id": "de3c77cecd_7", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 10.3}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 14.1}]}", "target": "bottle", "split": "train"}
{"id": "de3c77cecd_8", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "Yes", "target": "handbag", "split": "train"}
{"id": "de3c77cecd_9", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the handbag in this scene?\nA. file binder\nB. phone\nC. papers\nD. office chair", "choices": ["A. file binder", "B. phone", "C. papers", "D. office chair"], "answer": "D. office chair", "target": "handbag", "split": "train"}
{"id": "de3c77cecd_10", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the handbag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The handbag is to the side of and below the office chair, about 7 cm away. It is also to the side of and below the papers, about 44 cm away. It is further to the side of and below the file binder, about 44 cm away.", "target": "handbag", "split": "train"}
{"id": "de3c77cecd_11", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the handbag in structured JSON format.", "choices": null, "answer": "{\"target\": \"handbag\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 6.7}, {\"object\": \"papers\", \"relation\": \"side_below\", \"distance_cm\": 43.9}]}", "target": "handbag", "split": "train"}
{"id": "de3c77cecd_12", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "Yes", "target": "mouse pad", "split": "train"}
{"id": "de3c77cecd_13", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. books\nB. webcam\nC. hole punch\nD. office chair", "choices": ["A. books", "B. webcam", "C. hole punch", "D. office chair"], "answer": "D. office chair", "target": "mouse pad", "split": "train"}
{"id": "de3c77cecd_14", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the table. It is also to the side of and above the office chair, about 7 cm away. It is further near the hole punch, about 24 cm away.", "target": "mouse pad", "split": "train"}
{"id": "de3c77cecd_15", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 7.2}, {\"object\": \"hole punch\", \"relation\": \"near\", \"distance_cm\": 23.9}]}", "target": "mouse pad", "split": "train"}
{"id": "de3c77cecd_16", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glue bottle appear?", "choices": null, "answer": "No", "target": "glue bottle", "split": "train"}
{"id": "de3c77cecd_17", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone in this scene?\nA. divider\nB. cabinet\nC. mug\nD. file binder", "choices": ["A. divider", "B. cabinet", "C. mug", "D. file binder"], "answer": "D. file binder", "target": "phone", "split": "train"}
{"id": "de3c77cecd_18", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone is next to the file binder. It is also to the side of and above the cabinet, about 17 cm away. It is further near the mug, about 26 cm away.", "target": "phone", "split": "train"}
{"id": "de3c77cecd_19", "scene_id": "de3c77cecd", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"file binder\", \"relation\": \"next_to\", \"distance_cm\": 2.2}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 17.1}]}", "target": "phone", "split": "train"}
{"id": "de7dc9993f_0", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "de7dc9993f_1", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse pad\nB. monitor\nC. office chair\nD. mouse", "choices": ["A. mouse pad", "B. monitor", "C. office chair", "D. mouse"], "answer": "A. mouse pad", "target": "keyboard", "split": "train"}
{"id": "de7dc9993f_2", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the mouse pad, about 6 cm away. It is further to the side of and above the office chair, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "de7dc9993f_3", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse pad\", \"relation\": \"next_to\", \"distance_cm\": 6.0}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 9.5}]}", "target": "keyboard", "split": "train"}
{"id": "de7dc9993f_4", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "de7dc9993f_5", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. desk\nD. office chair", "choices": ["A. keyboard", "B. monitor", "C. desk", "D. office chair"], "answer": "C. desk", "target": "mouse", "split": "train"}
{"id": "de7dc9993f_6", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the desk, about 7 cm away. It is also to the side of and below the monitor, about 13 cm away. It is further next to the keyboard, about 13 cm away.", "target": "mouse", "split": "train"}
{"id": "de7dc9993f_7", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 6.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.8}]}", "target": "mouse", "split": "train"}
{"id": "de7dc9993f_8", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "Yes", "target": "mouse pad", "split": "train"}
{"id": "de7dc9993f_9", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. power strip\nB. keyboard\nC. office chair\nD. file cabinet", "choices": ["A. power strip", "B. keyboard", "C. office chair", "D. file cabinet"], "answer": "B. keyboard", "target": "mouse pad", "split": "train"}
{"id": "de7dc9993f_10", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the desk. It is also next to the keyboard, about 6 cm away. It is further to the side of and above the office chair, about 8 cm away.", "target": "mouse pad", "split": "train"}
{"id": "de7dc9993f_11", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.0}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 8.4}]}", "target": "mouse pad", "split": "train"}
{"id": "de7dc9993f_12", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphone case appear?", "choices": null, "answer": "No", "target": "headphone case", "split": "train"}
{"id": "de7dc9993f_13", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. mouse pad\nB. file cabinet\nC. cable\nD. mouse", "choices": ["A. mouse pad", "B. file cabinet", "C. cable", "D. mouse"], "answer": "A. mouse pad", "target": "power strip", "split": "train"}
{"id": "de7dc9993f_14", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the desk. It is also near the mouse pad, about 17 cm away. It is further to the side of and above the file cabinet, about 20 cm away.", "target": "power strip", "split": "train"}
{"id": "de7dc9993f_15", "scene_id": "de7dc9993f", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse pad\", \"relation\": \"near\", \"distance_cm\": 16.6}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 20.2}]}", "target": "power strip", "split": "train"}
{"id": "de83a9c2d7_0", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "de83a9c2d7_1", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. heater\nB. table\nC. chair\nD. mug", "choices": ["A. heater", "B. table", "C. chair", "D. mug"], "answer": "A. heater", "target": "extension cord", "split": "train"}
{"id": "de83a9c2d7_2", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the heater, about 35 cm away. It is also to the side of and below the chair, about 42 cm away.", "target": "extension cord", "split": "train"}
{"id": "de83a9c2d7_3", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 34.9}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 42.4}]}", "target": "extension cord", "split": "train"}
{"id": "de83a9c2d7_4", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "No", "target": "bottle", "split": "train"}
{"id": "de83a9c2d7_5", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. heater\nB. table\nC. chair\nD. extension cord", "choices": ["A. heater", "B. table", "C. chair", "D. extension cord"], "answer": "C. chair", "target": "mug", "split": "train"}
{"id": "de83a9c2d7_6", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and above the chair, about 50 cm away.", "target": "mug", "split": "train"}
{"id": "de83a9c2d7_7", "scene_id": "de83a9c2d7", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 50.2}]}", "target": "mug", "split": "train"}
{"id": "dfa70fb232_0", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "dfa70fb232_1", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. refrigerator\nB. poster\nC. oven\nD. kitchen cabinet", "choices": ["A. refrigerator", "B. poster", "C. oven", "D. kitchen cabinet"], "answer": "D. kitchen cabinet", "target": "bottle", "split": "train"}
{"id": "dfa70fb232_2", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the oven, about 46 cm away. It is also to the side of and above the refrigerator, about 46 cm away. It is further to the side of and below the poster, about 59 cm away.", "target": "bottle", "split": "train"}
{"id": "dfa70fb232_3", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"oven\", \"relation\": \"side_above\", \"distance_cm\": 45.9}, {\"object\": \"refrigerator\", \"relation\": \"side_above\", \"distance_cm\": 46.4}]}", "target": "bottle", "split": "train"}
{"id": "dfa70fb232_4", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "No", "target": "phone charger", "split": "train"}
{"id": "dfa70fb232_5", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. cutting board\nB. bench\nC. piano\nD. bottle", "choices": ["A. cutting board", "B. bench", "C. piano", "D. bottle"], "answer": "B. bench", "target": "power strip", "split": "train"}
{"id": "dfa70fb232_6", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also to the side of and above the bench, about 27 cm away. It is further to the side of and above the piano, about 28 cm away.", "target": "power strip", "split": "train"}
{"id": "dfa70fb232_7", "scene_id": "dfa70fb232", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bench\", \"relation\": \"side_above\", \"distance_cm\": 27.2}, {\"object\": \"piano\", \"relation\": \"side_above\", \"distance_cm\": 28.4}]}", "target": "power strip", "split": "train"}
{"id": "e050c15a8d_0", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "e050c15a8d_1", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. monitor\nB. pc\nC. keyabord\nD. mouse", "choices": ["A. monitor", "B. pc", "C. keyabord", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "train"}
{"id": "e050c15a8d_2", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 3 cm away. It is further near the keyabord, about 21 cm away.", "target": "keyboard", "split": "train"}
{"id": "e050c15a8d_3", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 2.9}, {\"object\": \"keyabord\", \"relation\": \"near\", \"distance_cm\": 20.9}]}", "target": "keyboard", "split": "train"}
{"id": "e050c15a8d_4", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "e050c15a8d_5", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. chair\nB. whiteboard eraser\nC. whiteboard\nD. box", "choices": ["A. chair", "B. whiteboard eraser", "C. whiteboard", "D. box"], "answer": "C. whiteboard", "target": "spray bottle", "split": "train"}
{"id": "e050c15a8d_6", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the table. It is also to the side of and below the whiteboard, about 9 cm away. It is further to the side of and above the whiteboard eraser, about 15 cm away.", "target": "spray bottle", "split": "train"}
{"id": "e050c15a8d_7", "scene_id": "e050c15a8d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"whiteboard\", \"relation\": \"side_below\", \"distance_cm\": 8.8}, {\"object\": \"whiteboard eraser\", \"relation\": \"side_above\", \"distance_cm\": 15.3}]}", "target": "spray bottle", "split": "train"}
{"id": "e0abd740ba_0", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "e0abd740ba_1", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. monitor\nB. cabinet\nC. desk lamp\nD. office chair", "choices": ["A. monitor", "B. cabinet", "C. desk lamp", "D. office chair"], "answer": "A. monitor", "target": "cable", "split": "train"}
{"id": "e0abd740ba_2", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and above the office chair, about 28 cm away. It is also to the side of and below the desk lamp, about 34 cm away. It is further to the side of and above the cabinet, about 39 cm away.", "target": "cable", "split": "train"}
{"id": "e0abd740ba_3", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 28.2}, {\"object\": \"desk lamp\", \"relation\": \"side_below\", \"distance_cm\": 34.0}]}", "target": "cable", "split": "train"}
{"id": "e0abd740ba_4", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "e0abd740ba_5", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. book\nB. cabinet\nC. desk lamp\nD. cable", "choices": ["A. book", "B. cabinet", "C. desk lamp", "D. cable"], "answer": "A. book", "target": "spray bottle", "split": "train"}
{"id": "e0abd740ba_6", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and above the book, about 55 cm away.", "target": "spray bottle", "split": "train"}
{"id": "e0abd740ba_7", "scene_id": "e0abd740ba", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"book\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 55.3}]}", "target": "spray bottle", "split": "train"}
{"id": "e1b1d9de55_0", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "No", "target": "cable", "split": "train"}
{"id": "e1b1d9de55_1", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. paper\nB. book\nC. hot bag\nD. monitor", "choices": ["A. paper", "B. book", "C. hot bag", "D. monitor"], "answer": "C. hot bag", "target": "keyboard", "split": "train"}
{"id": "e1b1d9de55_2", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the hot bag, about 7 cm away. It is further to the side of and below the monitor, about 19 cm away.", "target": "keyboard", "split": "train"}
{"id": "e1b1d9de55_3", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"hot bag\", \"relation\": \"next_to\", \"distance_cm\": 6.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 19.1}]}", "target": "keyboard", "split": "train"}
{"id": "e1b1d9de55_4", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "e1b1d9de55_5", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. envelope\nB. telephone\nC. rolled poster\nD. plant", "choices": ["A. envelope", "B. telephone", "C. rolled poster", "D. plant"], "answer": "C. rolled poster", "target": "power strip", "split": "train"}
{"id": "e1b1d9de55_6", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the desk. It is also next to the rolled poster, about 10 cm away. It is further near the envelope, about 30 cm away.", "target": "power strip", "split": "train"}
{"id": "e1b1d9de55_7", "scene_id": "e1b1d9de55", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"rolled poster\", \"relation\": \"next_to\", \"distance_cm\": 10.3}, {\"object\": \"envelope\", \"relation\": \"near\", \"distance_cm\": 30.3}]}", "target": "power strip", "split": "train"}
{"id": "e1fb8626c8_0", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "e1fb8626c8_1", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. cardboard box\nB. mouse\nC. office chair\nD. table lamp", "choices": ["A. cardboard box", "B. mouse", "C. office chair", "D. table lamp"], "answer": "A. cardboard box", "target": "headphones", "split": "train"}
{"id": "e1fb8626c8_2", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the desk. It is also next to the cardboard box, about 9 cm away. It is further near the mouse, about 22 cm away.", "target": "headphones", "split": "train"}
{"id": "e1fb8626c8_3", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"cardboard box\", \"relation\": \"next_to\", \"distance_cm\": 9.0}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 21.7}]}", "target": "headphones", "split": "train"}
{"id": "e1fb8626c8_4", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "e1fb8626c8_5", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. water bottle\nB. headphones\nC. table lamp\nD. notebook", "choices": ["A. water bottle", "B. headphones", "C. table lamp", "D. notebook"], "answer": "C. table lamp", "target": "mouse", "split": "train"}
{"id": "e1fb8626c8_6", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also to the side of and below the table lamp, about 9 cm away. It is further near the headphones, about 22 cm away.", "target": "mouse", "split": "train"}
{"id": "e1fb8626c8_7", "scene_id": "e1fb8626c8", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 8.6}, {\"object\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 21.7}]}", "target": "mouse", "split": "train"}
{"id": "e35232b92f_0", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "e35232b92f_1", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. chair\nB. book\nC. notebook\nD. laptop", "choices": ["A. chair", "B. book", "C. notebook", "D. laptop"], "answer": "D. laptop", "target": "bottle", "split": "train"}
{"id": "e35232b92f_2", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the laptop, about 12 cm away. It is further near the book, about 46 cm away.", "target": "bottle", "split": "train"}
{"id": "e35232b92f_3", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 11.9}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 46.0}]}", "target": "bottle", "split": "train"}
{"id": "e35232b92f_4", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "train"}
{"id": "e35232b92f_5", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. nightstand\nB. air humidifier\nC. book\nD. thermos", "choices": ["A. nightstand", "B. air humidifier", "C. book", "D. thermos"], "answer": "A. nightstand", "target": "laptop charger", "split": "train"}
{"id": "e35232b92f_6", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the nightstand, about 9 cm away. It is also to the side of and below the air humidifier, about 19 cm away. It is further to the side of and below the thermos, about 47 cm away.", "target": "laptop charger", "split": "train"}
{"id": "e35232b92f_7", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 9.1}, {\"object\": \"air humidifier\", \"relation\": \"side_below\", \"distance_cm\": 18.6}]}", "target": "laptop charger", "split": "train"}
{"id": "e35232b92f_8", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "e35232b92f_9", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. table\nB. chair\nC. laptop\nD. router", "choices": ["A. table", "B. chair", "C. laptop", "D. router"], "answer": "D. router", "target": "notebook", "split": "train"}
{"id": "e35232b92f_10", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and above the router, about 6 cm away. It is also to the side of and above the table, about 10 cm away. It is further to the side of and above the laptop, about 24 cm away.", "target": "notebook", "split": "train"}
{"id": "e35232b92f_11", "scene_id": "e35232b92f", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"router\", \"relation\": \"side_above\", \"distance_cm\": 5.9}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 10.1}]}", "target": "notebook", "split": "train"}
{"id": "e3ad7115db_0", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "train"}
{"id": "e3ad7115db_1", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. files\nB. papers\nC. rolling chair\nD. desk", "choices": ["A. files", "B. papers", "C. rolling chair", "D. desk"], "answer": "C. rolling chair", "target": "bottle", "split": "train"}
{"id": "e3ad7115db_2", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the rolling chair, about 12 cm away. It is also to the side of and below the desk, about 37 cm away. It is further to the side of and below the files, about 42 cm away.", "target": "bottle", "split": "train"}
{"id": "e3ad7115db_3", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"drawer\", \"references\": [{\"object\": \"rolling chair\", \"relation\": \"side_below\", \"distance_cm\": 12.2}, {\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 36.6}]}", "target": "bottle", "split": "train"}
{"id": "e3ad7115db_4", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "e3ad7115db_5", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. books\nB. book\nC. rolling chair\nD. papers", "choices": ["A. books", "B. book", "C. rolling chair", "D. papers"], "answer": "D. papers", "target": "cup", "split": "train"}
{"id": "e3ad7115db_6", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the desk. It is also next to the papers, about 5 cm away. It is further to the side of and above the rolling chair, about 7 cm away.", "target": "cup", "split": "train"}
{"id": "e3ad7115db_7", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 4.6}, {\"object\": \"rolling chair\", \"relation\": \"side_above\", \"distance_cm\": 7.3}]}", "target": "cup", "split": "train"}
{"id": "e3ad7115db_8", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "e3ad7115db_9", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. drawer\nB. power socket\nC. phone charger\nD. rolling chair", "choices": ["A. drawer", "B. power socket", "C. phone charger", "D. rolling chair"], "answer": "D. rolling chair", "target": "extension cord", "split": "train"}
{"id": "e3ad7115db_10", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the rolling chair, about 7 cm away. It is also to the side of and below the phone charger, about 52 cm away. It is further to the side of and below the power socket, about 53 cm away.", "target": "extension cord", "split": "train"}
{"id": "e3ad7115db_11", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"rolling chair\", \"relation\": \"side_below\", \"distance_cm\": 6.7}, {\"object\": \"phone charger\", \"relation\": \"side_below\", \"distance_cm\": 52.4}]}", "target": "extension cord", "split": "train"}
{"id": "e3ad7115db_12", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "e3ad7115db_13", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. rolling chair\nB. extension cord\nC. picture\nD. desk", "choices": ["A. rolling chair", "B. extension cord", "C. picture", "D. desk"], "answer": "D. desk", "target": "phone charger", "split": "train"}
{"id": "e3ad7115db_14", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the desk, about 27 cm away. It is also to the side of and below the rolling chair, about 30 cm away. It is further to the side of and above the extension cord, about 52 cm away.", "target": "phone charger", "split": "train"}
{"id": "e3ad7115db_15", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 26.8}, {\"object\": \"rolling chair\", \"relation\": \"side_below\", \"distance_cm\": 30.4}]}", "target": "phone charger", "split": "train"}
{"id": "e3ad7115db_16", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "e3ad7115db_17", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the sticky note in this scene?\nA. pen stand\nB. desk\nC. decoration\nD. pens", "choices": ["A. pen stand", "B. desk", "C. decoration", "D. pens"], "answer": "B. desk", "target": "sticky note", "split": "train"}
{"id": "e3ad7115db_18", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the sticky note in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The sticky note is to the side of and above the desk, about 3 cm away. It is also to the side of and below the pens, about 8 cm away. It is further next to the pen stand, about 8 cm away.", "target": "sticky note", "split": "train"}
{"id": "e3ad7115db_19", "scene_id": "e3ad7115db", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the sticky note in structured JSON format.", "choices": null, "answer": "{\"target\": \"sticky note\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 2.9}, {\"object\": \"pens\", \"relation\": \"side_below\", \"distance_cm\": 7.8}]}", "target": "sticky note", "split": "train"}
{"id": "e3c1da58dd_0", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "e3c1da58dd_1", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the laptop cover, about 23 cm away. It is also to the side of and below the box, about 24 cm away. It is further to the side of and below the books, about 27 cm away.", "target": "backpack", "split": "train"}
{"id": "e3c1da58dd_2", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"box\", \"references\": [{\"object\": \"laptop cover\", \"relation\": \"side_below\", \"distance_cm\": 23.0}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 23.5}]}", "target": "backpack", "split": "train"}
{"id": "e3c1da58dd_3", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "e3c1da58dd_4", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. poster\nB. box\nC. telephone\nD. laptop cover", "choices": ["A. poster", "B. box", "C. telephone", "D. laptop cover"], "answer": "D. laptop cover", "target": "headphones", "split": "train"}
{"id": "e3c1da58dd_5", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and below the laptop cover, about 6 cm away. It is also to the side of and below the poster, about 21 cm away. It is further to the side of and above the telephone, about 32 cm away.", "target": "headphones", "split": "train"}
{"id": "e3c1da58dd_6", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"box\", \"references\": [{\"object\": \"laptop cover\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"poster\", \"relation\": \"side_below\", \"distance_cm\": 20.6}]}", "target": "headphones", "split": "train"}
{"id": "e3c1da58dd_7", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "e3c1da58dd_8", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and below the monitor stand, about 13 cm away. It is also to the side of and below the books, about 13 cm away. It is further to the side of and below the laptop, about 14 cm away.", "target": "keyboard", "split": "train"}
{"id": "e3c1da58dd_9", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"monitor stand\", \"relation\": \"side_below\", \"distance_cm\": 12.8}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 13.5}]}", "target": "keyboard", "split": "train"}
{"id": "e3c1da58dd_10", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "e3c1da58dd_11", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. cabinet\nC. monitor stand\nD. keyboard", "choices": ["A. office chair", "B. cabinet", "C. monitor stand", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "e3c1da58dd_12", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 14 cm away. It is also to the side of and above the cabinet, about 17 cm away. It is further to the side of and above the office chair, about 20 cm away.", "target": "mouse", "split": "train"}
{"id": "e3c1da58dd_13", "scene_id": "e3c1da58dd", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 14.5}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 17.4}]}", "target": "mouse", "split": "train"}
{"id": "e3c39f4aea_0", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "e3c39f4aea_1", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. box\nB. picture\nC. bag\nD. books", "choices": ["A. box", "B. picture", "C. bag", "D. books"], "answer": "C. bag", "target": "bottle", "split": "train"}
{"id": "e3c39f4aea_2", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the cabinet. It is also to the side of and above the bag, about 14 cm away. It is further to the side of and below the picture, about 30 cm away.", "target": "bottle", "split": "train"}
{"id": "e3c39f4aea_3", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 13.8}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 30.2}]}", "target": "bottle", "split": "train"}
{"id": "e3c39f4aea_4", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "e3c39f4aea_5", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. office chair\nB. water jug\nC. package\nD. mouse", "choices": ["A. office chair", "B. water jug", "C. package", "D. mouse"], "answer": "B. water jug", "target": "cup", "split": "train"}
{"id": "e3c39f4aea_6", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also under the water jug, about 4 cm away. It is further next to the package, about 12 cm away.", "target": "cup", "split": "train"}
{"id": "e3c39f4aea_7", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"water jug\", \"relation\": \"under\", \"distance_cm\": 4.2}, {\"object\": \"package\", \"relation\": \"next_to\", \"distance_cm\": 11.9}]}", "target": "cup", "split": "train"}
{"id": "e3c39f4aea_8", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "train"}
{"id": "e3c39f4aea_9", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. table\nB. sofa\nC. box\nD. mug", "choices": ["A. table", "B. sofa", "C. box", "D. mug"], "answer": "B. sofa", "target": "headphones", "split": "train"}
{"id": "e3c39f4aea_10", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the sofa, about 4 cm away. It is also to the side of and above the box, about 18 cm away. It is further near the table, about 33 cm away.", "target": "headphones", "split": "train"}
{"id": "e3c39f4aea_11", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"sofa\", \"relation\": \"next_to\", \"distance_cm\": 3.8}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 17.8}]}", "target": "headphones", "split": "train"}
{"id": "e3c39f4aea_12", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tv remote appear?", "choices": null, "answer": "No", "target": "tv remote", "split": "train"}
{"id": "e3c39f4aea_13", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. chair\nB. mouse\nC. cup\nD. water jug", "choices": ["A. chair", "B. mouse", "C. cup", "D. water jug"], "answer": "B. mouse", "target": "keyboard", "split": "train"}
{"id": "e3c39f4aea_14", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 4 cm away. It is further near the cup, about 29 cm away.", "target": "keyboard", "split": "train"}
{"id": "e3c39f4aea_15", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.2}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 28.6}]}", "target": "keyboard", "split": "train"}
{"id": "e3c39f4aea_16", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "e3c39f4aea_17", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. laptop\nB. keyboard\nC. cup\nD. office chair", "choices": ["A. laptop", "B. keyboard", "C. cup", "D. office chair"], "answer": "B. keyboard", "target": "mouse", "split": "train"}
{"id": "e3c39f4aea_18", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 4 cm away. It is further to the side of and below the laptop, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "e3c39f4aea_19", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.2}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 9.5}]}", "target": "mouse", "split": "train"}
{"id": "e3c39f4aea_20", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "e3c39f4aea_21", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. computer tower\nB. printer\nC. paper shredder\nD. spray bottle", "choices": ["A. computer tower", "B. printer", "C. paper shredder", "D. spray bottle"], "answer": "B. printer", "target": "mug", "split": "train"}
{"id": "e3c39f4aea_22", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and below the printer, about 34 cm away. It is further near the spray bottle, about 40 cm away.", "target": "mug", "split": "train"}
{"id": "e3c39f4aea_23", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"printer\", \"relation\": \"side_below\", \"distance_cm\": 33.8}, {\"object\": \"spray bottle\", \"relation\": \"near\", \"distance_cm\": 39.9}]}", "target": "mug", "split": "train"}
{"id": "e3c39f4aea_24", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "Yes", "target": "tote bag", "split": "train"}
{"id": "e3c39f4aea_25", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. box\nB. nightstand\nC. cabinet\nD. power socket", "choices": ["A. box", "B. nightstand", "C. cabinet", "D. power socket"], "answer": "B. nightstand", "target": "tote bag", "split": "train"}
{"id": "e3c39f4aea_26", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is under the nightstand, about 4 cm away. It is also to the side of and below the cabinet, about 40 cm away. It is further to the side of and below the power socket, about 44 cm away.", "target": "tote bag", "split": "train"}
{"id": "e3c39f4aea_27", "scene_id": "e3c39f4aea", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"nightstand\", \"references\": [{\"object\": \"nightstand\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 39.8}]}", "target": "tote bag", "split": "train"}
{"id": "e3e0617f98_0", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "e3e0617f98_1", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. blinds\nB. bathroom wall\nC. bathtub\nD. book", "choices": ["A. blinds", "B. bathroom wall", "C. bathtub", "D. book"], "answer": "B. bathroom wall", "target": "backpack", "split": "train"}
{"id": "e3e0617f98_2", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the bathroom wall, about 29 cm away. It is also to the side of and above the rug, about 32 cm away. It is further to the side of and below the door, about 43 cm away.", "target": "backpack", "split": "train"}
{"id": "e3e0617f98_3", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"rug\", \"references\": [{\"object\": \"bathroom wall\", \"relation\": \"side_below\", \"distance_cm\": 29.1}, {\"object\": \"rug\", \"relation\": \"side_above\", \"distance_cm\": 32.2}]}", "target": "backpack", "split": "train"}
{"id": "e3e0617f98_4", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "train"}
{"id": "e3e0617f98_5", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. kitchen counter\nB. jar\nC. box\nD. windowsill", "choices": ["A. kitchen counter", "B. jar", "C. box", "D. windowsill"], "answer": "A. kitchen counter", "target": "bottle", "split": "train"}
{"id": "e3e0617f98_6", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the kitchen counter, about 7 cm away. It is also to the side of and below the windowsill, about 10 cm away. It is further next to the box, about 10 cm away.", "target": "bottle", "split": "train"}
{"id": "e3e0617f98_7", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 6.5}, {\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 9.8}]}", "target": "bottle", "split": "train"}
{"id": "e3e0617f98_8", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "e3e0617f98_9", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the book, about 9 cm away. It is further next to the mouse, about 9 cm away.", "target": "keyboard", "split": "train"}
{"id": "e3e0617f98_10", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 8.9}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 9.3}]}", "target": "keyboard", "split": "train"}
{"id": "e3e0617f98_11", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "e3e0617f98_12", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. office chair\nD. book", "choices": ["A. keyboard", "B. monitor", "C. office chair", "D. book"], "answer": "D. book", "target": "mouse", "split": "train"}
{"id": "e3e0617f98_13", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the book, about 6 cm away. It is further next to the keyboard, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "e3e0617f98_14", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 5.6}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.3}]}", "target": "mouse", "split": "train"}
{"id": "e3e0617f98_15", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone appear?", "choices": null, "answer": "No", "target": "phone", "split": "train"}
{"id": "e3e0617f98_16", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the table. It is also next to the coaster, about 3 cm away. It is further next to the paper sheets, about 4 cm away.", "target": "pen", "split": "train"}
{"id": "e3e0617f98_17", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"coaster\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"paper sheets\", \"relation\": \"next_to\", \"distance_cm\": 3.6}]}", "target": "pen", "split": "train"}
{"id": "e3e0617f98_18", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "e3e0617f98_19", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "Yes", "target": "soap", "split": "train"}
{"id": "e3e0617f98_20", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap in this scene?\nA. backpack\nB. towel\nC. faucet\nD. pipe", "choices": ["A. backpack", "B. towel", "C. faucet", "D. pipe"], "answer": "D. pipe", "target": "soap", "split": "train"}
{"id": "e3e0617f98_21", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap is to the side of and below the pipe, about 58 cm away. It is also to the side of and below the faucet, about 64 cm away. It is further to the side of and above the towel, about 72 cm away.", "target": "soap", "split": "train"}
{"id": "e3e0617f98_22", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap\", \"support_surface\": \"bathroom wall\", \"references\": [{\"object\": \"pipe\", \"relation\": \"side_below\", \"distance_cm\": 57.9}, {\"object\": \"faucet\", \"relation\": \"side_below\", \"distance_cm\": 64.5}]}", "target": "soap", "split": "train"}
{"id": "e3e0617f98_23", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "e3e0617f98_24", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tv remote in this scene?\nA. power sockets\nB. bathroom wall\nC. backpack\nD. television", "choices": ["A. power sockets", "B. bathroom wall", "C. backpack", "D. television"], "answer": "D. television", "target": "tv remote", "split": "train"}
{"id": "e3e0617f98_25", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is on the cabinet. It is also to the side of and below the television, about 13 cm away. It is further to the side of and below the power sockets, about 79 cm away.", "target": "tv remote", "split": "train"}
{"id": "e3e0617f98_26", "scene_id": "e3e0617f98", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"television\", \"relation\": \"side_below\", \"distance_cm\": 13.3}, {\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 79.1}]}", "target": "tv remote", "split": "train"}
{"id": "e3ecd49e2b_0", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mobile phone appear?", "choices": null, "answer": "No", "target": "mobile phone", "split": "train"}
{"id": "e3ecd49e2b_1", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. speaker\nC. mouse\nD. windowsill", "choices": ["A. monitor", "B. speaker", "C. mouse", "D. windowsill"], "answer": "D. windowsill", "target": "bottle", "split": "train"}
{"id": "e3ecd49e2b_2", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is near the windowsill, about 21 cm away. It is also to the side of and below the monitor, about 24 cm away. It is further near the speaker, about 28 cm away.", "target": "bottle", "split": "train"}
{"id": "e3ecd49e2b_3", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"rubber water bag\", \"references\": [{\"object\": \"windowsill\", \"relation\": \"near\", \"distance_cm\": 20.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 23.9}]}", "target": "bottle", "split": "train"}
{"id": "e3ecd49e2b_4", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "Yes", "target": "eraser", "split": "train"}
{"id": "e3ecd49e2b_5", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the eraser in this scene?\nA. whiteboard eraser\nB. office chair\nC. jacket\nD. whiteboard marker", "choices": ["A. whiteboard eraser", "B. office chair", "C. jacket", "D. whiteboard marker"], "answer": "D. whiteboard marker", "target": "eraser", "split": "train"}
{"id": "e3ecd49e2b_6", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the eraser in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The eraser is next to the whiteboard marker, about 3 cm away. It is also to the side of and above the office chair, about 26 cm away. It is further near the whiteboard eraser, about 26 cm away.", "target": "eraser", "split": "train"}
{"id": "e3ecd49e2b_7", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the eraser in structured JSON format.", "choices": null, "answer": "{\"target\": \"eraser\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"whiteboard marker\", \"relation\": \"next_to\", \"distance_cm\": 3.5}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 25.7}]}", "target": "eraser", "split": "train"}
{"id": "e3ecd49e2b_8", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "Yes", "target": "headset", "split": "train"}
{"id": "e3ecd49e2b_9", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headset in this scene?\nA. box\nB. laptop stand\nC. office chair\nD. keyboard", "choices": ["A. box", "B. laptop stand", "C. office chair", "D. keyboard"], "answer": "A. box", "target": "headset", "split": "train"}
{"id": "e3ecd49e2b_10", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is to the side of and below the box, about 12 cm away. It is also next to the laptop stand, about 15 cm away. It is further near the keyboard, about 18 cm away.", "target": "headset", "split": "train"}
{"id": "e3ecd49e2b_11", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"laptop stand\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 12.2}, {\"object\": \"laptop stand\", \"relation\": \"next_to\", \"distance_cm\": 14.7}]}", "target": "headset", "split": "train"}
{"id": "e3ecd49e2b_12", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "e3ecd49e2b_13", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. monitor\nB. keyboard\nC. mouse\nD. bag", "choices": ["A. monitor", "B. keyboard", "C. mouse", "D. bag"], "answer": "A. monitor", "target": "power strip", "split": "train"}
{"id": "e3ecd49e2b_14", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also next to the bag, about 13 cm away. It is further next to the keyboard, about 14 cm away.", "target": "power strip", "split": "train"}
{"id": "e3ecd49e2b_15", "scene_id": "e3ecd49e2b", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 12.8}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 13.6}]}", "target": "power strip", "split": "train"}
{"id": "e4e625a3e4_0", "scene_id": "e4e625a3e4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "e4e625a3e4_1", "scene_id": "e4e625a3e4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. sink\nB. walking sticks\nC. bags\nD. shoes", "choices": ["A. sink", "B. walking sticks", "C. bags", "D. shoes"], "answer": "D. shoes", "target": "backpack", "split": "train"}
{"id": "e4e625a3e4_2", "scene_id": "e4e625a3e4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the shoes, about 3 cm away. It is also to the side of and below the walking sticks, about 48 cm away. It is further to the side of and below the sink, about 61 cm away.", "target": "backpack", "split": "train"}
{"id": "e4e625a3e4_3", "scene_id": "e4e625a3e4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"shoes\", \"references\": [{\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"walking sticks\", \"relation\": \"side_below\", \"distance_cm\": 47.9}]}", "target": "backpack", "split": "train"}
{"id": "e4fb2a623b_0", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "e4fb2a623b_1", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bag\nB. heater\nC. book\nD. mouse", "choices": ["A. bag", "B. heater", "C. book", "D. mouse"], "answer": "D. mouse", "target": "backpack", "split": "train"}
{"id": "e4fb2a623b_2", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the mouse, about 41 cm away. It is also to the side of and below the heater, about 54 cm away.", "target": "backpack", "split": "train"}
{"id": "e4fb2a623b_3", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"side_below\", \"distance_cm\": 40.8}, {\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 54.0}]}", "target": "backpack", "split": "train"}
{"id": "e4fb2a623b_4", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "e4fb2a623b_5", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. counter\nC. wires\nD. extension board", "choices": ["A. monitor", "B. counter", "C. wires", "D. extension board"], "answer": "B. counter", "target": "bottle", "split": "train"}
{"id": "e4fb2a623b_6", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the counter, about 3 cm away. It is also to the side of and above the extension board, about 20 cm away. It is further to the side of and above the wires, about 32 cm away.", "target": "bottle", "split": "train"}
{"id": "e4fb2a623b_7", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"counter\", \"relation\": \"side_above\", \"distance_cm\": 2.8}, {\"object\": \"extension board\", \"relation\": \"side_above\", \"distance_cm\": 20.1}]}", "target": "bottle", "split": "train"}
{"id": "e4fb2a623b_8", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "e4fb2a623b_9", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. book\nB. bag\nC. monitor\nD. extension board", "choices": ["A. book", "B. bag", "C. monitor", "D. extension board"], "answer": "A. book", "target": "cup", "split": "train"}
{"id": "e4fb2a623b_10", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the book, about 7 cm away. It is also near the extension board, about 44 cm away. It is further to the side of and below the monitor, about 51 cm away.", "target": "cup", "split": "train"}
{"id": "e4fb2a623b_11", "scene_id": "e4fb2a623b", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 7.2}, {\"object\": \"extension board\", \"relation\": \"near\", \"distance_cm\": 44.4}]}", "target": "cup", "split": "train"}
{"id": "e7ac609391_0", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "e7ac609391_1", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and above the chair, about 7 cm away. It is further next to the mug, about 8 cm away.", "target": "keyboard", "split": "train"}
{"id": "e7ac609391_2", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 7.2}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 7.7}]}", "target": "keyboard", "split": "train"}
{"id": "e7ac609391_3", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "e7ac609391_4", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. wardrobe\nC. suitcase\nD. keyboard", "choices": ["A. chair", "B. wardrobe", "C. suitcase", "D. keyboard"], "answer": "C. suitcase", "target": "mouse", "split": "train"}
{"id": "e7ac609391_5", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also to the side of and above the suitcase, about 18 cm away. It is further to the side of and above the chair, about 23 cm away.", "target": "mouse", "split": "train"}
{"id": "e7ac609391_6", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"suitcase\", \"relation\": \"side_above\", \"distance_cm\": 18.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 23.0}]}", "target": "mouse", "split": "train"}
{"id": "e7ac609391_7", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "e7ac609391_8", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. milk carton\nB. keyboard\nC. chair\nD. monitor", "choices": ["A. milk carton", "B. keyboard", "C. chair", "D. monitor"], "answer": "B. keyboard", "target": "mug", "split": "train"}
{"id": "e7ac609391_9", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the keyboard, about 8 cm away. It is further to the side of and below the milk carton, about 16 cm away.", "target": "mug", "split": "train"}
{"id": "e7ac609391_10", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.7}, {\"object\": \"milk carton\", \"relation\": \"side_below\", \"distance_cm\": 15.7}]}", "target": "mug", "split": "train"}
{"id": "e7ac609391_11", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "e7ac609391_12", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scissors in this scene?\nA. spice rack\nB. oven\nC. pot holder\nD. stove", "choices": ["A. spice rack", "B. oven", "C. pot holder", "D. stove"], "answer": "C. pot holder", "target": "scissors", "split": "train"}
{"id": "e7ac609391_13", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is under the pot holder. It is also to the side of and above the stove, about 23 cm away. It is further below the spice rack, about 26 cm away.", "target": "scissors", "split": "train"}
{"id": "e7ac609391_14", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"spice rack\", \"references\": [{\"object\": \"pot holder\", \"relation\": \"under\", \"distance_cm\": 2.4}, {\"object\": \"stove\", \"relation\": \"side_above\", \"distance_cm\": 23.4}]}", "target": "scissors", "split": "train"}
{"id": "e7ac609391_15", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "Yes", "target": "tissue", "split": "train"}
{"id": "e7ac609391_16", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue in this scene?\nA. toilet paper\nB. chair\nC. laptop stand\nD. laptop", "choices": ["A. toilet paper", "B. chair", "C. laptop stand", "D. laptop"], "answer": "A. toilet paper", "target": "tissue", "split": "train"}
{"id": "e7ac609391_17", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue is on the table. It is also next to the toilet paper, about 5 cm away. It is further to the side of and above the chair, about 12 cm away.", "target": "tissue", "split": "train"}
{"id": "e7ac609391_18", "scene_id": "e7ac609391", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue\", \"support_surface\": \"table\", \"references\": [{\"object\": \"toilet paper\", \"relation\": \"next_to\", \"distance_cm\": 4.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 11.9}]}", "target": "tissue", "split": "train"}
{"id": "e7af285f7d_0", "scene_id": "e7af285f7d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "train"}
{"id": "e7af285f7d_1", "scene_id": "e7af285f7d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. toothpaste\nB. mirror\nC. washbasin\nD. towel", "choices": ["A. toothpaste", "B. mirror", "C. washbasin", "D. towel"], "answer": "A. toothpaste", "target": "bottle", "split": "train"}
{"id": "e7af285f7d_2", "scene_id": "e7af285f7d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the cabinet. It is also to the side of and above the toothpaste, about 5 cm away. It is further to the side of and below the mirror, about 21 cm away.", "target": "bottle", "split": "train"}
{"id": "e7af285f7d_3", "scene_id": "e7af285f7d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"toothpaste\", \"relation\": \"side_above\", \"distance_cm\": 5.3}, {\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 21.4}]}", "target": "bottle", "split": "train"}
{"id": "e7ccd75e5d_0", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "e7ccd75e5d_1", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. books\nB. spectacles case\nC. headphones\nD. mug", "choices": ["A. books", "B. spectacles case", "C. headphones", "D. mug"], "answer": "B. spectacles case", "target": "bottle", "split": "train"}
{"id": "e7ccd75e5d_2", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the spectacles case, about 5 cm away. It is further next to the headphones, about 9 cm away.", "target": "bottle", "split": "train"}
{"id": "e7ccd75e5d_3", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"spectacles case\", \"relation\": \"side_above\", \"distance_cm\": 4.9}, {\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 8.9}]}", "target": "bottle", "split": "train"}
{"id": "e7ccd75e5d_4", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "e7ccd75e5d_5", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. bottle\nB. paper towel\nC. laptop\nD. spectacles case", "choices": ["A. bottle", "B. paper towel", "C. laptop", "D. spectacles case"], "answer": "D. spectacles case", "target": "headphones", "split": "train"}
{"id": "e7ccd75e5d_6", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also next to the spectacles case, about 5 cm away. It is further next to the bottle, about 9 cm away.", "target": "headphones", "split": "train"}
{"id": "e7ccd75e5d_7", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"spectacles case\", \"relation\": \"next_to\", \"distance_cm\": 5.3}, {\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 8.9}]}", "target": "headphones", "split": "train"}
{"id": "e7ccd75e5d_8", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "e7ccd75e5d_9", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. paper towel\nB. bottle\nC. spectacles case\nD. headphones", "choices": ["A. paper towel", "B. bottle", "C. spectacles case", "D. headphones"], "answer": "C. spectacles case", "target": "mug", "split": "train"}
{"id": "e7ccd75e5d_10", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the spectacles case, about 13 cm away. It is further to the side of and below the paper towel, about 15 cm away.", "target": "mug", "split": "train"}
{"id": "e7ccd75e5d_11", "scene_id": "e7ccd75e5d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"spectacles case\", \"relation\": \"next_to\", \"distance_cm\": 13.4}, {\"object\": \"paper towel\", \"relation\": \"side_below\", \"distance_cm\": 15.1}]}", "target": "mug", "split": "train"}
{"id": "e81c8b3eec_0", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the deodorant bottle appear?", "choices": null, "answer": "No", "target": "deodorant bottle", "split": "train"}
{"id": "e81c8b3eec_1", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. kitchen cabinet\nB. power socket\nC. bag\nD. light switch", "choices": ["A. kitchen cabinet", "B. power socket", "C. bag", "D. light switch"], "answer": "A. kitchen cabinet", "target": "cup", "split": "train"}
{"id": "e81c8b3eec_2", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the kitchen cabinet, about 3 cm away. It is also to the side of and below the light switch, about 14 cm away. It is further to the side of and below the door, about 37 cm away.", "target": "cup", "split": "train"}
{"id": "e81c8b3eec_3", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"kitchen cabinet\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 13.9}]}", "target": "cup", "split": "train"}
{"id": "e81c8b3eec_4", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "e81c8b3eec_5", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. basket\nB. windowsill\nC. window frame\nD. water bottle", "choices": ["A. basket", "B. windowsill", "C. window frame", "D. water bottle"], "answer": "D. water bottle", "target": "mug", "split": "train"}
{"id": "e81c8b3eec_6", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and below the water bottle, about 7 cm away. It is further to the side of and below the window frame, about 20 cm away.", "target": "mug", "split": "train"}
{"id": "e81c8b3eec_7", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"water bottle\", \"relation\": \"side_below\", \"distance_cm\": 7.5}, {\"object\": \"window frame\", \"relation\": \"side_below\", \"distance_cm\": 20.4}]}", "target": "mug", "split": "train"}
{"id": "e81c8b3eec_8", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "train"}
{"id": "e81c8b3eec_9", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. mug\nB. windowsill\nC. table\nD. window frame", "choices": ["A. mug", "B. windowsill", "C. table", "D. window frame"], "answer": "C. table", "target": "water bottle", "split": "train"}
{"id": "e81c8b3eec_10", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is to the side of and above the table. It is also to the side of and above the mug, about 7 cm away. It is further to the side of and below the window frame, about 22 cm away.", "target": "water bottle", "split": "train"}
{"id": "e81c8b3eec_11", "scene_id": "e81c8b3eec", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.0}, {\"object\": \"mug\", \"relation\": \"side_above\", \"distance_cm\": 7.5}]}", "target": "water bottle", "split": "train"}
{"id": "e86c6742b9_0", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "Yes", "target": "power adapter", "split": "train"}
{"id": "e86c6742b9_1", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. table\nB. decoration\nC. chair\nD. wallet", "choices": ["A. table", "B. decoration", "C. chair", "D. wallet"], "answer": "A. table", "target": "power adapter", "split": "train"}
{"id": "e86c6742b9_2", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and below the table, about 16 cm away. It is also to the side of and below the wallet, about 52 cm away. It is further to the side of and below the decoration, about 67 cm away.", "target": "power adapter", "split": "train"}
{"id": "e86c6742b9_3", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 16.1}, {\"object\": \"wallet\", \"relation\": \"side_below\", \"distance_cm\": 51.9}]}", "target": "power adapter", "split": "train"}
{"id": "e86c6742b9_4", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "e86c6742b9_5", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the wallet in this scene?\nA. power sockets\nB. window frame\nC. heater\nD. decoration", "choices": ["A. power sockets", "B. window frame", "C. heater", "D. decoration"], "answer": "D. decoration", "target": "wallet", "split": "train"}
{"id": "e86c6742b9_6", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the wallet in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The wallet is on the table. It is also to the side of and below the decoration, about 18 cm away. It is further to the side of and below the window frame, about 35 cm away.", "target": "wallet", "split": "train"}
{"id": "e86c6742b9_7", "scene_id": "e86c6742b9", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the wallet in structured JSON format.", "choices": null, "answer": "{\"target\": \"wallet\", \"support_surface\": \"table\", \"references\": [{\"object\": \"decoration\", \"relation\": \"side_below\", \"distance_cm\": 18.4}, {\"object\": \"window frame\", \"relation\": \"side_below\", \"distance_cm\": 35.0}]}", "target": "wallet", "split": "train"}
{"id": "e87f1ebcac_0", "scene_id": "e87f1ebcac", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "e87f1ebcac_1", "scene_id": "e87f1ebcac", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. shower wall\nB. towel rack\nC. shower\nD. light", "choices": ["A. shower wall", "B. towel rack", "C. shower", "D. light"], "answer": "A. shower wall", "target": "bottle", "split": "train"}
{"id": "e87f1ebcac_2", "scene_id": "e87f1ebcac", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the shower wall, about 14 cm away. It is also to the side of and above the shower, about 43 cm away. It is further to the side of and below the light, about 50 cm away.", "target": "bottle", "split": "train"}
{"id": "e87f1ebcac_3", "scene_id": "e87f1ebcac", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"towel rack\", \"references\": [{\"object\": \"shower wall\", \"relation\": \"side_above\", \"distance_cm\": 14.2}, {\"object\": \"shower\", \"relation\": \"side_above\", \"distance_cm\": 43.3}]}", "target": "bottle", "split": "train"}
{"id": "e898c76c1f_0", "scene_id": "e898c76c1f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "No", "target": "handbag", "split": "train"}
{"id": "e898c76c1f_1", "scene_id": "e898c76c1f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone in this scene?\nA. table\nB. power socket\nC. chair\nD. cabinet", "choices": ["A. table", "B. power socket", "C. chair", "D. cabinet"], "answer": "D. cabinet", "target": "phone", "split": "train"}
{"id": "e898c76c1f_2", "scene_id": "e898c76c1f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone is on the table. It is also to the side of and above the cabinet, about 14 cm away. It is further to the side of and above the chair, about 39 cm away.", "target": "phone", "split": "train"}
{"id": "e898c76c1f_3", "scene_id": "e898c76c1f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 14.0}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 39.5}]}", "target": "phone", "split": "train"}
{"id": "e8e81396b6_0", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "e8e81396b6_1", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. office chair\nB. monitor\nC. window sill\nD. telephone", "choices": ["A. office chair", "B. monitor", "C. window sill", "D. telephone"], "answer": "D. telephone", "target": "bottle", "split": "train"}
{"id": "e8e81396b6_2", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the telephone, about 8 cm away. It is also to the side of and below the monitor, about 32 cm away. It is further to the side of and above the window sill, about 39 cm away.", "target": "bottle", "split": "train"}
{"id": "e8e81396b6_3", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"window sill\", \"references\": [{\"object\": \"telephone\", \"relation\": \"side_above\", \"distance_cm\": 7.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 32.3}]}", "target": "bottle", "split": "train"}
{"id": "e8e81396b6_4", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "e8e81396b6_5", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. mug\nB. monitor\nC. bottle\nD. keyboard", "choices": ["A. mug", "B. monitor", "C. bottle", "D. keyboard"], "answer": "D. keyboard", "target": "mouse", "split": "train"}
{"id": "e8e81396b6_6", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 5 cm away. It is also next to the mug, about 9 cm away. It is further to the side of and below the monitor, about 30 cm away.", "target": "mouse", "split": "train"}
{"id": "e8e81396b6_7", "scene_id": "e8e81396b6", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"keyboard\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.7}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 9.5}]}", "target": "mouse", "split": "train"}
{"id": "e9ac2fc517_0", "scene_id": "e9ac2fc517", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "e9ac2fc517_1", "scene_id": "e9ac2fc517", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. keyboard\nB. table\nC. office chair\nD. monitor", "choices": ["A. keyboard", "B. table", "C. office chair", "D. monitor"], "answer": "C. office chair", "target": "backpack", "split": "train"}
{"id": "e9ac2fc517_2", "scene_id": "e9ac2fc517", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the table. It is also to the side of and above the office chair, about 22 cm away.", "target": "backpack", "split": "train"}
{"id": "e9ac2fc517_3", "scene_id": "e9ac2fc517", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 22.0}]}", "target": "backpack", "split": "train"}
{"id": "e9e16b6043_0", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "e9e16b6043_1", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is next to the watering pot, about 12 cm away. It is also to the side of and below the plant, about 13 cm away. It is further to the side of and below the office chair, about 17 cm away.", "target": "backpack", "split": "train"}
{"id": "e9e16b6043_2", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"watering pot\", \"relation\": \"next_to\", \"distance_cm\": 12.4}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 13.2}]}", "target": "backpack", "split": "train"}
{"id": "e9e16b6043_3", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "e9e16b6043_4", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. monitor\nC. office chair\nD. laptop stand", "choices": ["A. mouse", "B. monitor", "C. office chair", "D. laptop stand"], "answer": "A. mouse", "target": "keyboard", "split": "train"}
{"id": "e9e16b6043_5", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also next to the mouse, about 4 cm away. It is further to the side of and above the office chair, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "e9e16b6043_6", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 10.1}]}", "target": "keyboard", "split": "train"}
{"id": "e9e16b6043_7", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "e9e16b6043_8", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. table\nD. office chair", "choices": ["A. keyboard", "B. monitor", "C. table", "D. office chair"], "answer": "A. keyboard", "target": "mouse", "split": "train"}
{"id": "e9e16b6043_9", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 4 cm away. It is also to the side of and above the table, about 6 cm away. It is further to the side of and above the office chair, about 13 cm away.", "target": "mouse", "split": "train"}
{"id": "e9e16b6043_10", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.3}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 6.0}]}", "target": "mouse", "split": "train"}
{"id": "e9e16b6043_11", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the hairbrush appear?", "choices": null, "answer": "No", "target": "hairbrush", "split": "train"}
{"id": "e9e16b6043_12", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse pad in this scene?\nA. sticky note\nB. office chair\nC. laptop stand\nD. monitor", "choices": ["A. sticky note", "B. office chair", "C. laptop stand", "D. monitor"], "answer": "B. office chair", "target": "mouse pad", "split": "train"}
{"id": "e9e16b6043_13", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is on the table. It is also to the side of and above the office chair, about 7 cm away. It is further to the side of and below the monitor, about 13 cm away.", "target": "mouse pad", "split": "train"}
{"id": "e9e16b6043_14", "scene_id": "e9e16b6043", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 7.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.7}]}", "target": "mouse pad", "split": "train"}
{"id": "ea3b3710a2_0", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "train"}
{"id": "ea3b3710a2_1", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. heater\nB. book\nC. panel\nD. bedside table", "choices": ["A. heater", "B. book", "C. panel", "D. bedside table"], "answer": "A. heater", "target": "backpack", "split": "train"}
{"id": "ea3b3710a2_2", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the heater, about 7 cm away. It is also to the side of and below the panel, about 64 cm away.", "target": "backpack", "split": "train"}
{"id": "ea3b3710a2_3", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"heater\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 7.3}, {\"object\": \"panel\", \"relation\": \"side_below\", \"distance_cm\": 63.6}]}", "target": "backpack", "split": "train"}
{"id": "ea3b3710a2_4", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "ea3b3710a2_5", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is next to the box, about 3 cm away. It is also above the shelf, about 3 cm away. It is further to the side of and above the clothes, about 6 cm away.", "target": "bottle", "split": "train"}
{"id": "ea3b3710a2_6", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"shelf\", \"relation\": \"above\", \"distance_cm\": 3.4}]}", "target": "bottle", "split": "train"}
{"id": "ea3b3710a2_7", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "Yes", "target": "glasses case", "split": "train"}
{"id": "ea3b3710a2_8", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the glasses case in this scene?\nA. egg carton\nB. shelf\nC. bottle\nD. monitor", "choices": ["A. egg carton", "B. shelf", "C. bottle", "D. monitor"], "answer": "B. shelf", "target": "glasses case", "split": "train"}
{"id": "ea3b3710a2_9", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the glasses case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The glasses case is to the side of and above the monitor, about 18 cm away. It is also to the side of and below the bottle, about 19 cm away. It is further near the egg carton, about 39 cm away.", "target": "glasses case", "split": "train"}
{"id": "ea3b3710a2_10", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the glasses case in structured JSON format.", "choices": null, "answer": "{\"target\": \"glasses case\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_above\", \"distance_cm\": 17.9}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 18.5}]}", "target": "glasses case", "split": "train"}
{"id": "ea3b3710a2_11", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "ea3b3710a2_12", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. can\nB. water bottle\nC. book\nD. table", "choices": ["A. can", "B. water bottle", "C. book", "D. table"], "answer": "C. book", "target": "mouse", "split": "train"}
{"id": "ea3b3710a2_13", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the table, about 8 cm away. It is also to the side of and below the water bottle, about 14 cm away. It is further to the side of and below the can, about 29 cm away.", "target": "mouse", "split": "train"}
{"id": "ea3b3710a2_14", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 7.8}, {\"object\": \"water bottle\", \"relation\": \"side_below\", \"distance_cm\": 13.9}]}", "target": "mouse", "split": "train"}
{"id": "ea3b3710a2_15", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "train"}
{"id": "ea3b3710a2_16", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. bedside table\nB. book\nC. monitor\nD. tissue box", "choices": ["A. bedside table", "B. book", "C. monitor", "D. tissue box"], "answer": "B. book", "target": "power adapter", "split": "train"}
{"id": "ea3b3710a2_17", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is on the table. It is also near the book, about 23 cm away. It is further to the side of and below the monitor, about 32 cm away.", "target": "power adapter", "split": "train"}
{"id": "ea3b3710a2_18", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 23.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 32.1}]}", "target": "power adapter", "split": "train"}
{"id": "ea3b3710a2_19", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wallet appear?", "choices": null, "answer": "No", "target": "wallet", "split": "train"}
{"id": "ea3b3710a2_20", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. book\nB. pillow\nC. tissue box\nD. power adapter", "choices": ["A. book", "B. pillow", "C. tissue box", "D. power adapter"], "answer": "C. tissue box", "target": "power strip", "split": "train"}
{"id": "ea3b3710a2_21", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is next to the tissue box, about 13 cm away. It is also to the side of and below the pillow, about 27 cm away. It is further to the side of and below the power adapter, about 48 cm away.", "target": "power strip", "split": "train"}
{"id": "ea3b3710a2_22", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"tissue box\", \"relation\": \"next_to\", \"distance_cm\": 13.0}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 26.6}]}", "target": "power strip", "split": "train"}
{"id": "ea3b3710a2_23", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "train"}
{"id": "ea3b3710a2_24", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. power strip\nB. book\nC. speaker\nD. table", "choices": ["A. power strip", "B. book", "C. speaker", "D. table"], "answer": "C. speaker", "target": "tissue box", "split": "train"}
{"id": "ea3b3710a2_25", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is next to the speaker, about 3 cm away. It is also next to the power strip, about 13 cm away. It is further to the side of and below the table, about 18 cm away.", "target": "tissue box", "split": "train"}
{"id": "ea3b3710a2_26", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"speaker\", \"relation\": \"next_to\", \"distance_cm\": 3.5}, {\"object\": \"power strip\", \"relation\": \"next_to\", \"distance_cm\": 13.0}]}", "target": "tissue box", "split": "train"}
{"id": "ea3b3710a2_27", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the handbag appear?", "choices": null, "answer": "No", "target": "handbag", "split": "train"}
{"id": "ea3b3710a2_28", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. monitor\nB. book\nC. mouse\nD. can", "choices": ["A. monitor", "B. book", "C. mouse", "D. can"], "answer": "B. book", "target": "water bottle", "split": "train"}
{"id": "ea3b3710a2_29", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the table. It is also to the side of and above the book, about 4 cm away. It is further next to the can, about 9 cm away.", "target": "water bottle", "split": "train"}
{"id": "ea3b3710a2_30", "scene_id": "ea3b3710a2", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 4.3}, {\"object\": \"can\", \"relation\": \"next_to\", \"distance_cm\": 9.4}]}", "target": "water bottle", "split": "train"}
{"id": "ea42cd27e6_0", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "ea42cd27e6_1", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is next to the can, about 5 cm away. It is also to the side of and above the blanket, about 11 cm away. It is further above the light switch, about 42 cm away.", "target": "bottle", "split": "train"}
{"id": "ea42cd27e6_2", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"can\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"blanket\", \"relation\": \"side_above\", \"distance_cm\": 11.4}]}", "target": "bottle", "split": "train"}
{"id": "ea42cd27e6_3", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "ea42cd27e6_4", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. ironing board\nB. weight plate\nC. pillow\nD. books", "choices": ["A. ironing board", "B. weight plate", "C. pillow", "D. books"], "answer": "A. ironing board", "target": "extension cord", "split": "train"}
{"id": "ea42cd27e6_5", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is next to the ironing board, about 7 cm away. It is also to the side of and below the weight plate, about 37 cm away. It is further to the side of and below the books, about 40 cm away.", "target": "extension cord", "split": "train"}
{"id": "ea42cd27e6_6", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"pillow\", \"references\": [{\"object\": \"ironing board\", \"relation\": \"next_to\", \"distance_cm\": 6.6}, {\"object\": \"weight plate\", \"relation\": \"side_below\", \"distance_cm\": 36.8}]}", "target": "extension cord", "split": "train"}
{"id": "ea42cd27e6_7", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "ea42cd27e6_8", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. suitcase\nB. trash can\nC. router\nD. weight plates", "choices": ["A. suitcase", "B. trash can", "C. router", "D. weight plates"], "answer": "B. trash can", "target": "laptop charger", "split": "train"}
{"id": "ea42cd27e6_9", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the trash can, about 20 cm away. It is also to the side of and below the suitcase, about 26 cm away. It is further to the side of and below the weight plates, about 26 cm away.", "target": "laptop charger", "split": "train"}
{"id": "ea42cd27e6_10", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 20.3}, {\"object\": \"suitcase\", \"relation\": \"side_below\", \"distance_cm\": 25.9}]}", "target": "laptop charger", "split": "train"}
{"id": "ea42cd27e6_11", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "ea42cd27e6_12", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. chair\nB. blanket\nC. monitor\nD. plastic bag", "choices": ["A. chair", "B. blanket", "C. monitor", "D. plastic bag"], "answer": "D. plastic bag", "target": "mouse", "split": "train"}
{"id": "ea42cd27e6_13", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is under the plastic bag, about 3 cm away. It is also to the side of and above the chair, about 19 cm away. It is further to the side of and below the monitor, about 23 cm away.", "target": "mouse", "split": "train"}
{"id": "ea42cd27e6_14", "scene_id": "ea42cd27e6", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"plastic bag\", \"relation\": \"under\", \"distance_cm\": 2.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 19.3}]}", "target": "mouse", "split": "train"}
{"id": "eaa6c90310_0", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "eaa6c90310_1", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. cabinet\nB. water pitcher\nC. pot\nD. extension cord", "choices": ["A. cabinet", "B. water pitcher", "C. pot", "D. extension cord"], "answer": "C. pot", "target": "bottle", "split": "train"}
{"id": "eaa6c90310_2", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also to the side of and above the pot, about 9 cm away. It is further above the extension cord, about 18 cm away.", "target": "bottle", "split": "train"}
{"id": "eaa6c90310_3", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"pot\", \"relation\": \"side_above\", \"distance_cm\": 9.2}, {\"object\": \"extension cord\", \"relation\": \"above\", \"distance_cm\": 17.5}]}", "target": "bottle", "split": "train"}
{"id": "eaa6c90310_4", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "eaa6c90310_5", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. bottle\nB. desk lamp\nC. shelf\nD. pot", "choices": ["A. bottle", "B. desk lamp", "C. shelf", "D. pot"], "answer": "C. shelf", "target": "extension cord", "split": "train"}
{"id": "eaa6c90310_6", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is on the cabinet. It is also to the side of and below the shelf, about 13 cm away. It is further below the bottle, about 18 cm away.", "target": "extension cord", "split": "train"}
{"id": "eaa6c90310_7", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 12.5}, {\"object\": \"bottle\", \"relation\": \"below\", \"distance_cm\": 17.5}]}", "target": "extension cord", "split": "train"}
{"id": "eaa6c90310_8", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "eaa6c90310_9", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. desk\nB. pedestal fan\nC. cabinet\nD. chair", "choices": ["A. desk", "B. pedestal fan", "C. cabinet", "D. chair"], "answer": "D. chair", "target": "laptop charger", "split": "train"}
{"id": "eaa6c90310_10", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is to the side of and below the chair, about 10 cm away. It is also to the side of and below the desk, about 11 cm away. It is further to the side of and below the pedestal fan, about 18 cm away.", "target": "laptop charger", "split": "train"}
{"id": "eaa6c90310_11", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 9.7}, {\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 11.1}]}", "target": "laptop charger", "split": "train"}
{"id": "eaa6c90310_12", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "eaa6c90310_13", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. cup\nB. chair\nC. luggage\nD. laptop bag", "choices": ["A. cup", "B. chair", "C. luggage", "D. laptop bag"], "answer": "D. laptop bag", "target": "mouse", "split": "train"}
{"id": "eaa6c90310_14", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also to the side of and below the laptop bag, about 5 cm away. It is further next to the cup, about 7 cm away.", "target": "mouse", "split": "train"}
{"id": "eaa6c90310_15", "scene_id": "eaa6c90310", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop bag\", \"relation\": \"side_below\", \"distance_cm\": 5.1}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "mouse", "split": "train"}
{"id": "eaab7bcc15_0", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "train"}
{"id": "eaab7bcc15_1", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. pouch\nB. light switch\nC. bag\nD. power socket", "choices": ["A. pouch", "B. light switch", "C. bag", "D. power socket"], "answer": "A. pouch", "target": "backpack", "split": "train"}
{"id": "eaab7bcc15_2", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the pouch, about 18 cm away. It is also to the side of and below the bag, about 34 cm away. It is further to the side of and above the power socket, about 44 cm away.", "target": "backpack", "split": "train"}
{"id": "eaab7bcc15_3", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"pouch\", \"relation\": \"side_below\", \"distance_cm\": 17.6}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 34.0}]}", "target": "backpack", "split": "train"}
{"id": "eaab7bcc15_4", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "eaab7bcc15_5", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. documents\nB. document folder\nC. backpack\nD. bag", "choices": ["A. documents", "B. document folder", "C. backpack", "D. bag"], "answer": "B. document folder", "target": "cup", "split": "train"}
{"id": "eaab7bcc15_6", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the document folder, about 17 cm away. It is also to the side of and above the bag, about 20 cm away. It is further to the side of and above the documents, about 35 cm away.", "target": "cup", "split": "train"}
{"id": "eaab7bcc15_7", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"document folder\", \"relation\": \"side_above\", \"distance_cm\": 17.4}, {\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 19.9}]}", "target": "cup", "split": "train"}
{"id": "eaab7bcc15_8", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "train"}
{"id": "eaab7bcc15_9", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. boots\nB. papers\nC. paper\nD. plant pot", "choices": ["A. boots", "B. papers", "C. paper", "D. plant pot"], "answer": "A. boots", "target": "mug", "split": "train"}
{"id": "eaab7bcc15_10", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also to the side of and above the boots, about 33 cm away. It is further to the side of and above the paper, about 42 cm away.", "target": "mug", "split": "train"}
{"id": "eaab7bcc15_11", "scene_id": "eaab7bcc15", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"boots\", \"relation\": \"side_above\", \"distance_cm\": 33.2}, {\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 41.5}]}", "target": "mug", "split": "train"}
{"id": "eab5494dca_0", "scene_id": "eab5494dca", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "eab5494dca_1", "scene_id": "eab5494dca", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the shelf. It is also next to the plates. It is further next to the pot, about 3 cm away.", "target": "mug", "split": "train"}
{"id": "eab5494dca_2", "scene_id": "eab5494dca", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"plates\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"pot\", \"relation\": \"next_to\", \"distance_cm\": 3.1}]}", "target": "mug", "split": "train"}
{"id": "eb8ef9b4cc_0", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil appear?", "choices": null, "answer": "No", "target": "pencil", "split": "train"}
{"id": "eb8ef9b4cc_1", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. desk\nB. cooking mittens\nC. floor lamp\nD. cardboard box", "choices": ["A. desk", "B. cooking mittens", "C. floor lamp", "D. cardboard box"], "answer": "B. cooking mittens", "target": "bottle", "split": "train"}
{"id": "eb8ef9b4cc_2", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is next to the cooking mittens, about 5 cm away. It is also next to the cardboard box, about 13 cm away. It is further to the side of and above the floor lamp, about 16 cm away.", "target": "bottle", "split": "train"}
{"id": "eb8ef9b4cc_3", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cooking mittens\", \"relation\": \"next_to\", \"distance_cm\": 5.0}, {\"object\": \"cardboard box\", \"relation\": \"next_to\", \"distance_cm\": 13.1}]}", "target": "bottle", "split": "train"}
{"id": "eb8ef9b4cc_4", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "eb8ef9b4cc_5", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the charger in this scene?\nA. pillow\nB. wooden panel window\nC. bed\nD. floor lamp", "choices": ["A. pillow", "B. wooden panel window", "C. bed", "D. floor lamp"], "answer": "D. floor lamp", "target": "charger", "split": "train"}
{"id": "eb8ef9b4cc_6", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is to the side of and below the floor lamp, about 11 cm away. It is also to the side of and below the wooden panel window, about 18 cm away. It is further to the side of and below the pillow, about 46 cm away.", "target": "charger", "split": "train"}
{"id": "eb8ef9b4cc_7", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"floor lamp\", \"references\": [{\"object\": \"floor lamp\", \"relation\": \"side_below\", \"distance_cm\": 11.2}, {\"object\": \"wooden panel window\", \"relation\": \"side_below\", \"distance_cm\": 18.1}]}", "target": "charger", "split": "train"}
{"id": "eb8ef9b4cc_8", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the lanyard appear?", "choices": null, "answer": "No", "target": "lanyard", "split": "train"}
{"id": "eb8ef9b4cc_9", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. pillow\nB. wooden panel window\nC. floor lamp\nD. bed", "choices": ["A. pillow", "B. wooden panel window", "C. floor lamp", "D. bed"], "answer": "C. floor lamp", "target": "extension cord", "split": "train"}
{"id": "eb8ef9b4cc_10", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the floor lamp, about 6 cm away. It is also to the side of and below the wooden panel window, about 17 cm away. It is further to the side of and below the pillow, about 51 cm away.", "target": "extension cord", "split": "train"}
{"id": "eb8ef9b4cc_11", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"floor lamp\", \"references\": [{\"object\": \"floor lamp\", \"relation\": \"side_below\", \"distance_cm\": 6.2}, {\"object\": \"wooden panel window\", \"relation\": \"side_below\", \"distance_cm\": 17.2}]}", "target": "extension cord", "split": "train"}
{"id": "eb8ef9b4cc_12", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "Yes", "target": "notepad", "split": "train"}
{"id": "eb8ef9b4cc_13", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notepad in this scene?\nA. bed\nB. bottle\nC. pillow\nD. floor lamp", "choices": ["A. bed", "B. bottle", "C. pillow", "D. floor lamp"], "answer": "A. bed", "target": "notepad", "split": "train"}
{"id": "eb8ef9b4cc_14", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notepad is next to the bed, about 3 cm away. It is also to the side of and below the pillow, about 14 cm away. It is further to the side of and above the floor lamp, about 78 cm away.", "target": "notepad", "split": "train"}
{"id": "eb8ef9b4cc_15", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"notepad\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"bed\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 14.0}]}", "target": "notepad", "split": "train"}
{"id": "eb8ef9b4cc_16", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "Yes", "target": "pencil case", "split": "train"}
{"id": "eb8ef9b4cc_17", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil case in this scene?\nA. pillow\nB. floor lamp\nC. bed\nD. bottle", "choices": ["A. pillow", "B. floor lamp", "C. bed", "D. bottle"], "answer": "A. pillow", "target": "pencil case", "split": "train"}
{"id": "eb8ef9b4cc_18", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil case in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil case is on the bed. It is also to the side of and below the pillow, about 14 cm away. It is further to the side of and above the floor lamp, about 86 cm away.", "target": "pencil case", "split": "train"}
{"id": "eb8ef9b4cc_19", "scene_id": "eb8ef9b4cc", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 14.3}, {\"object\": \"floor lamp\", \"relation\": \"side_above\", \"distance_cm\": 86.1}]}", "target": "pencil case", "split": "train"}
{"id": "ec614d4e30_0", "scene_id": "ec614d4e30", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "ec614d4e30_1", "scene_id": "ec614d4e30", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. table\nB. cable duct\nC. luggage\nD. bag", "choices": ["A. table", "B. cable duct", "C. luggage", "D. bag"], "answer": "A. table", "target": "bottle", "split": "train"}
{"id": "ec614d4e30_2", "scene_id": "ec614d4e30", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also to the side of and above the bag, about 17 cm away. It is further to the side of and below the cable duct, about 23 cm away.", "target": "bottle", "split": "train"}
{"id": "ec614d4e30_3", "scene_id": "ec614d4e30", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 17.3}, {\"object\": \"cable duct\", \"relation\": \"side_below\", \"distance_cm\": 22.7}]}", "target": "bottle", "split": "train"}
{"id": "ec614d4e30_4", "scene_id": "ec614d4e30", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "ec614d4e30_5", "scene_id": "ec614d4e30", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the door, about 10 cm away. It is also to the side of and below the chair, about 30 cm away. It is further to the side of and below the door window, about 30 cm away.", "target": "phone charger", "split": "train"}
{"id": "ec614d4e30_6", "scene_id": "ec614d4e30", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"door\", \"references\": [{\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 9.8}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 29.6}]}", "target": "phone charger", "split": "train"}
{"id": "ec9ff6051d_0", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "ec9ff6051d_1", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the backpag, about 4 cm away. It is also under the clothes, about 5 cm away. It is further to the side of and below the bag, about 17 cm away.", "target": "backpack", "split": "train"}
{"id": "ec9ff6051d_2", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"backpag\", \"relation\": \"side_above\", \"distance_cm\": 4.3}, {\"object\": \"clothes\", \"relation\": \"under\", \"distance_cm\": 4.8}]}", "target": "backpack", "split": "train"}
{"id": "ec9ff6051d_3", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "train"}
{"id": "ec9ff6051d_4", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and above the desk, about 4 cm away. It is also to the side of and above the magazines, about 4 cm away. It is further next to the files, about 9 cm away.", "target": "headphones", "split": "train"}
{"id": "ec9ff6051d_5", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"magazines\", \"relation\": \"side_above\", \"distance_cm\": 4.2}]}", "target": "headphones", "split": "train"}
{"id": "ec9ff6051d_6", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses appear?", "choices": null, "answer": "No", "target": "glasses", "split": "train"}
{"id": "ec9ff6051d_7", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. box\nB. laptop\nC. files\nD. plate", "choices": ["A. box", "B. laptop", "C. files", "D. plate"], "answer": "B. laptop", "target": "mouse", "split": "train"}
{"id": "ec9ff6051d_8", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also to the side of and below the laptop, about 7 cm away. It is further near the box, about 22 cm away.", "target": "mouse", "split": "train"}
{"id": "ec9ff6051d_9", "scene_id": "ec9ff6051d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 6.7}, {\"object\": \"box\", \"relation\": \"near\", \"distance_cm\": 22.2}]}", "target": "mouse", "split": "train"}
{"id": "eeeb9836b8_0", "scene_id": "eeeb9836b8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "train"}
{"id": "eeeb9836b8_1", "scene_id": "eeeb9836b8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. chair\nB. laptop\nC. window frame\nD. table", "choices": ["A. chair", "B. laptop", "C. window frame", "D. table"], "answer": "D. table", "target": "bottle", "split": "train"}
{"id": "eeeb9836b8_2", "scene_id": "eeeb9836b8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the table, about 3 cm away. It is also to the side of and above the laptop, about 7 cm away. It is further to the side of and below the window frame, about 26 cm away.", "target": "bottle", "split": "train"}
{"id": "eeeb9836b8_3", "scene_id": "eeeb9836b8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.6}, {\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 7.3}]}", "target": "bottle", "split": "train"}
{"id": "ef25276c25_0", "scene_id": "ef25276c25", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "No", "target": "mouse", "split": "train"}
{"id": "ef25276c25_1", "scene_id": "ef25276c25", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. table\nB. telephone\nC. chair\nD. power socket", "choices": ["A. table", "B. telephone", "C. chair", "D. power socket"], "answer": "D. power socket", "target": "power strip", "split": "train"}
{"id": "ef25276c25_2", "scene_id": "ef25276c25", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and above the power socket, about 7 cm away. It is also to the side of and above the telephone, about 10 cm away. It is further to the side of and above the table, about 20 cm away.", "target": "power strip", "split": "train"}
{"id": "ef25276c25_3", "scene_id": "ef25276c25", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 7.2}, {\"object\": \"telephone\", \"relation\": \"side_above\", \"distance_cm\": 10.4}]}", "target": "power strip", "split": "train"}
{"id": "ef60db4e8c_0", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "Yes", "target": "marker", "split": "train"}
{"id": "ef60db4e8c_1", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the marker in this scene?\nA. book\nB. table\nC. paper\nD. monitor", "choices": ["A. book", "B. table", "C. paper", "D. monitor"], "answer": "B. table", "target": "marker", "split": "train"}
{"id": "ef60db4e8c_2", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the marker in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The marker is to the side of and above the table, about 5 cm away. It is also to the side of and below the monitor, about 25 cm away. It is further near the book, about 31 cm away.", "target": "marker", "split": "train"}
{"id": "ef60db4e8c_3", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the marker in structured JSON format.", "choices": null, "answer": "{\"target\": \"marker\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 24.9}]}", "target": "marker", "split": "train"}
{"id": "ef60db4e8c_4", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "train"}
{"id": "ef60db4e8c_5", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. box\nB. book\nC. office chair\nD. picture", "choices": ["A. box", "B. book", "C. office chair", "D. picture"], "answer": "D. picture", "target": "power strip", "split": "train"}
{"id": "ef60db4e8c_6", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also to the side of and below the picture, about 31 cm away. It is further near the book, about 32 cm away.", "target": "power strip", "split": "train"}
{"id": "ef60db4e8c_7", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 30.8}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 32.3}]}", "target": "power strip", "split": "train"}
{"id": "ef60db4e8c_8", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "train"}
{"id": "ef60db4e8c_9", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the scissors in this scene?\nA. monitor\nB. book\nC. monitor stand\nD. office chair", "choices": ["A. monitor", "B. book", "C. monitor stand", "D. office chair"], "answer": "B. book", "target": "scissors", "split": "train"}
{"id": "ef60db4e8c_10", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the scissors in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scissors is on the table. It is also next to the book, about 14 cm away. It is further to the side of and below the monitor, about 17 cm away.", "target": "scissors", "split": "train"}
{"id": "ef60db4e8c_11", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the scissors in structured JSON format.", "choices": null, "answer": "{\"target\": \"scissors\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 13.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 17.2}]}", "target": "scissors", "split": "train"}
{"id": "ef60db4e8c_12", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the razor appear?", "choices": null, "answer": "No", "target": "razor", "split": "train"}
{"id": "ef60db4e8c_13", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the sticky note in this scene?\nA. table\nB. bottle\nC. paper\nD. keyboard", "choices": ["A. table", "B. bottle", "C. paper", "D. keyboard"], "answer": "A. table", "target": "sticky note", "split": "train"}
{"id": "ef60db4e8c_14", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the sticky note in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The sticky note is to the side of and above the table, about 5 cm away. It is also to the side of and above the keyboard, about 6 cm away. It is further to the side of and above the paper, about 19 cm away.", "target": "sticky note", "split": "train"}
{"id": "ef60db4e8c_15", "scene_id": "ef60db4e8c", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the sticky note in structured JSON format.", "choices": null, "answer": "{\"target\": \"sticky note\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.9}, {\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 6.1}]}", "target": "sticky note", "split": "train"}
{"id": "ef69d58016_0", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "ef69d58016_1", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. storage box\nB. bagpack\nC. chair\nD. cardboard box", "choices": ["A. storage box", "B. bagpack", "C. chair", "D. cardboard box"], "answer": "A. storage box", "target": "cup", "split": "train"}
{"id": "ef69d58016_2", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is on the table. It is also to the side of and above the storage box, about 19 cm away. It is further near the chair, about 25 cm away.", "target": "cup", "split": "train"}
{"id": "ef69d58016_3", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"storage box\", \"relation\": \"side_above\", \"distance_cm\": 18.5}, {\"object\": \"chair\", \"relation\": \"near\", \"distance_cm\": 25.3}]}", "target": "cup", "split": "train"}
{"id": "ef69d58016_4", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "train"}
{"id": "ef69d58016_5", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the handbag in this scene?\nA. pillow\nB. office chair\nC. mouse\nD. office table", "choices": ["A. pillow", "B. office chair", "C. mouse", "D. office table"], "answer": "D. office table", "target": "handbag", "split": "train"}
{"id": "ef69d58016_6", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the handbag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The handbag is to the side of and below the office table, about 7 cm away. It is also to the side of and below the office chair, about 32 cm away. It is further to the side of and below the pillow, about 41 cm away.", "target": "handbag", "split": "train"}
{"id": "ef69d58016_7", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the handbag in structured JSON format.", "choices": null, "answer": "{\"target\": \"handbag\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_below\", \"distance_cm\": 7.5}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 31.7}]}", "target": "handbag", "split": "train"}
{"id": "ef69d58016_8", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "ef69d58016_9", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. midi controller\nB. monitor\nC. office chair\nD. mouse", "choices": ["A. midi controller", "B. monitor", "C. office chair", "D. mouse"], "answer": "B. monitor", "target": "keyboard", "split": "train"}
{"id": "ef69d58016_10", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is under the monitor, about 3 cm away. It is also near the midi controller, about 21 cm away. It is further to the side of and above the office chair, about 25 cm away.", "target": "keyboard", "split": "train"}
{"id": "ef69d58016_11", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.8}, {\"object\": \"midi controller\", \"relation\": \"near\", \"distance_cm\": 21.3}]}", "target": "keyboard", "split": "train"}
{"id": "ef69d58016_12", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "ef69d58016_13", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. cabinet\nB. laptop sleeve\nC. pencil holder\nD. monitor", "choices": ["A. cabinet", "B. laptop sleeve", "C. pencil holder", "D. monitor"], "answer": "D. monitor", "target": "notebook", "split": "train"}
{"id": "ef69d58016_14", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and below the monitor, about 8 cm away. It is also to the side of and above the laptop sleeve, about 10 cm away. It is further next to the pencil holder, about 12 cm away.", "target": "notebook", "split": "train"}
{"id": "ef69d58016_15", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.8}, {\"object\": \"laptop sleeve\", \"relation\": \"side_above\", \"distance_cm\": 9.7}]}", "target": "notebook", "split": "train"}
{"id": "ef69d58016_16", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "Yes", "target": "pencil holder", "split": "train"}
{"id": "ef69d58016_17", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pencil holder in this scene?\nA. table lamp\nB. laptop sleeve\nC. cupboard\nD. notebook", "choices": ["A. table lamp", "B. laptop sleeve", "C. cupboard", "D. notebook"], "answer": "B. laptop sleeve", "target": "pencil holder", "split": "train"}
{"id": "ef69d58016_18", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pencil holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil holder is on the cabinet. It is also to the side of and above the laptop sleeve, about 9 cm away. It is further to the side of and below the table lamp, about 11 cm away.", "target": "pencil holder", "split": "train"}
{"id": "ef69d58016_19", "scene_id": "ef69d58016", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pencil holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil holder\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"laptop sleeve\", \"relation\": \"side_above\", \"distance_cm\": 9.0}, {\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 10.6}]}", "target": "pencil holder", "split": "train"}
{"id": "f07340dfea_0", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "f07340dfea_1", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. plug\nB. remote\nC. power strip\nD. chair", "choices": ["A. plug", "B. remote", "C. power strip", "D. chair"], "answer": "D. chair", "target": "notebook", "split": "train"}
{"id": "f07340dfea_2", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is to the side of and above the chair, about 41 cm away. It is also near the power strip, about 57 cm away. It is further to the side of and below the plug, about 71 cm away.", "target": "notebook", "split": "train"}
{"id": "f07340dfea_3", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 40.8}, {\"object\": \"power strip\", \"relation\": \"near\", \"distance_cm\": 57.0}]}", "target": "notebook", "split": "train"}
{"id": "f07340dfea_4", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "f07340dfea_5", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. notebook\nB. plug\nC. chair\nD. remote", "choices": ["A. notebook", "B. plug", "C. chair", "D. remote"], "answer": "C. chair", "target": "power strip", "split": "train"}
{"id": "f07340dfea_6", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is on the table. It is also to the side of and above the chair, about 48 cm away. It is further near the remote, about 51 cm away.", "target": "power strip", "split": "train"}
{"id": "f07340dfea_7", "scene_id": "f07340dfea", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 48.3}, {\"object\": \"remote\", \"relation\": \"near\", \"distance_cm\": 50.5}]}", "target": "power strip", "split": "train"}
{"id": "f1e01af60a_0", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "f1e01af60a_1", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. extension cord\nB. mouse\nC. laptop\nD. laptop bag", "choices": ["A. extension cord", "B. mouse", "C. laptop", "D. laptop bag"], "answer": "A. extension cord", "target": "backpack", "split": "train"}
{"id": "f1e01af60a_2", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and above the extension cord, about 28 cm away. It is also to the side of and below the laptop, about 33 cm away. It is further to the side of and below the laptop bag, about 36 cm away.", "target": "backpack", "split": "train"}
{"id": "f1e01af60a_3", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"extension cord\", \"relation\": \"side_above\", \"distance_cm\": 28.5}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 32.7}]}", "target": "backpack", "split": "train"}
{"id": "f1e01af60a_4", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue box appear?", "choices": null, "answer": "No", "target": "tissue box", "split": "train"}
{"id": "f1e01af60a_5", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. monitor\nB. book\nC. shelf\nD. teddy bear", "choices": ["A. monitor", "B. book", "C. shelf", "D. teddy bear"], "answer": "C. shelf", "target": "bottle", "split": "train"}
{"id": "f1e01af60a_6", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the shelf, about 3 cm away. It is also near the book, about 28 cm away. It is further to the side of and above the teddy bear, about 38 cm away.", "target": "bottle", "split": "train"}
{"id": "f1e01af60a_7", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 2.9}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 27.6}]}", "target": "bottle", "split": "train"}
{"id": "f1e01af60a_8", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "f1e01af60a_9", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. books\nB. mouse\nC. notebooks\nD. desk", "choices": ["A. books", "B. mouse", "C. notebooks", "D. desk"], "answer": "D. desk", "target": "keyboard", "split": "train"}
{"id": "f1e01af60a_10", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the desk, about 10 cm away. It is also near the notebooks, about 16 cm away. It is further near the mouse, about 21 cm away.", "target": "keyboard", "split": "train"}
{"id": "f1e01af60a_11", "scene_id": "f1e01af60a", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 10.4}, {\"object\": \"notebooks\", \"relation\": \"near\", \"distance_cm\": 16.3}]}", "target": "keyboard", "split": "train"}
{"id": "f1efd25854_0", "scene_id": "f1efd25854", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "train"}
{"id": "f1efd25854_1", "scene_id": "f1efd25854", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. soap bottle\nB. paper towel\nC. mirror\nD. paper towel dispenser", "choices": ["A. soap bottle", "B. paper towel", "C. mirror", "D. paper towel dispenser"], "answer": "C. mirror", "target": "bottle", "split": "train"}
{"id": "f1efd25854_2", "scene_id": "f1efd25854", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the mirror, about 14 cm away. It is also near the paper towel dispenser, about 38 cm away. It is further to the side of and above the soap bottle, about 42 cm away.", "target": "bottle", "split": "train"}
{"id": "f1efd25854_3", "scene_id": "f1efd25854", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"paper towel dispenser\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 13.6}, {\"object\": \"paper towel dispenser\", \"relation\": \"near\", \"distance_cm\": 38.1}]}", "target": "bottle", "split": "train"}
{"id": "f20ccd91de_0", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "f20ccd91de_1", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. bowl\nB. bottle\nC. bouquet\nD. book", "choices": ["A. bowl", "B. bottle", "C. bouquet", "D. book"], "answer": "D. book", "target": "backpack", "split": "train"}
{"id": "f20ccd91de_2", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the book, about 47 cm away. It is also to the side of and below the bottle, about 62 cm away.", "target": "backpack", "split": "train"}
{"id": "f20ccd91de_3", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"window frame\", \"references\": [{\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 47.1}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 62.5}]}", "target": "backpack", "split": "train"}
{"id": "f20ccd91de_4", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notepad appear?", "choices": null, "answer": "No", "target": "notepad", "split": "train"}
{"id": "f20ccd91de_5", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the handbag in this scene?\nA. picture frame\nB. notebook\nC. plush toy\nD. shelf rail", "choices": ["A. picture frame", "B. notebook", "C. plush toy", "D. shelf rail"], "answer": "D. shelf rail", "target": "handbag", "split": "train"}
{"id": "f20ccd91de_6", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the handbag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The handbag is on the shelf. It is also to the side of and above the shelf rail. It is further next to the plush toy, about 5 cm away.", "target": "handbag", "split": "train"}
{"id": "f20ccd91de_7", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the handbag in structured JSON format.", "choices": null, "answer": "{\"target\": \"handbag\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_above\", \"distance_cm\": 2.2}, {\"object\": \"plush toy\", \"relation\": \"next_to\", \"distance_cm\": 5.4}]}", "target": "handbag", "split": "train"}
{"id": "f20ccd91de_8", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "f20ccd91de_9", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. bouquet\nB. decoration\nC. bottle\nD. shelf rail", "choices": ["A. bouquet", "B. decoration", "C. bottle", "D. shelf rail"], "answer": "D. shelf rail", "target": "pen", "split": "train"}
{"id": "f20ccd91de_10", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the desk. It is also to the side of and below the shelf rail, about 12 cm away. It is further to the side of and below the decoration, about 15 cm away.", "target": "pen", "split": "train"}
{"id": "f20ccd91de_11", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 12.2}, {\"object\": \"decoration\", \"relation\": \"side_below\", \"distance_cm\": 14.6}]}", "target": "pen", "split": "train"}
{"id": "f20ccd91de_12", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "f20ccd91de_13", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tote bag in this scene?\nA. doorframe\nB. light switches\nC. towel\nD. umbrella", "choices": ["A. doorframe", "B. light switches", "C. towel", "D. umbrella"], "answer": "A. doorframe", "target": "tote bag", "split": "train"}
{"id": "f20ccd91de_14", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is attached to the door. It is also to the side of and above the doorframe, about 11 cm away. It is further to the side of and above the towel, about 13 cm away.", "target": "tote bag", "split": "train"}
{"id": "f20ccd91de_15", "scene_id": "f20ccd91de", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"door\", \"references\": [{\"object\": \"doorframe\", \"relation\": \"side_above\", \"distance_cm\": 10.6}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 12.6}]}", "target": "tote bag", "split": "train"}
{"id": "f248c2bcdc_0", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "f248c2bcdc_1", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. microscope\nB. mouse\nC. office chair\nD. tissue box", "choices": ["A. microscope", "B. mouse", "C. office chair", "D. tissue box"], "answer": "B. mouse", "target": "keyboard", "split": "train"}
{"id": "f248c2bcdc_2", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse, about 10 cm away. It is also to the side of and above the office chair, about 12 cm away. It is further near the tissue box, about 23 cm away.", "target": "keyboard", "split": "train"}
{"id": "f248c2bcdc_3", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 9.9}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.0}]}", "target": "keyboard", "split": "train"}
{"id": "f248c2bcdc_4", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "train"}
{"id": "f248c2bcdc_5", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. office chair\nC. monitor\nD. binder", "choices": ["A. keyboard", "B. office chair", "C. monitor", "D. binder"], "answer": "A. keyboard", "target": "mouse", "split": "train"}
{"id": "f248c2bcdc_6", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is next to the keyboard, about 10 cm away. It is also next to the binder, about 12 cm away. It is further to the side of and above the office chair, about 19 cm away.", "target": "mouse", "split": "train"}
{"id": "f248c2bcdc_7", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.9}, {\"object\": \"binder\", \"relation\": \"next_to\", \"distance_cm\": 11.9}]}", "target": "mouse", "split": "train"}
{"id": "f248c2bcdc_8", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "f248c2bcdc_9", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. monitor\nB. tissue box\nC. binder\nD. office chair", "choices": ["A. monitor", "B. tissue box", "C. binder", "D. office chair"], "answer": "C. binder", "target": "mousepad", "split": "train"}
{"id": "f248c2bcdc_10", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is next to the binder, about 7 cm away. It is also to the side of and above the office chair, about 12 cm away. It is further near the tissue box, about 20 cm away.", "target": "mousepad", "split": "train"}
{"id": "f248c2bcdc_11", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"binder\", \"relation\": \"next_to\", \"distance_cm\": 7.3}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.2}]}", "target": "mousepad", "split": "train"}
{"id": "f248c2bcdc_12", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "f248c2bcdc_13", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. machine button\nB. chemical bottle\nC. counter\nD. laboratory equipment", "choices": ["A. machine button", "B. chemical bottle", "C. counter", "D. laboratory equipment"], "answer": "D. laboratory equipment", "target": "spray bottle", "split": "train"}
{"id": "f248c2bcdc_14", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is on the counter. It is also to the side of and below the laboratory equipment, about 11 cm away. It is further to the side of and above the machine button, about 22 cm away.", "target": "spray bottle", "split": "train"}
{"id": "f248c2bcdc_15", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"laboratory equipment\", \"relation\": \"side_below\", \"distance_cm\": 10.7}, {\"object\": \"machine button\", \"relation\": \"side_above\", \"distance_cm\": 22.2}]}", "target": "spray bottle", "split": "train"}
{"id": "f248c2bcdc_16", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "train"}
{"id": "f248c2bcdc_17", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. keyboard\nB. mouse\nC. binder\nD. mousepad", "choices": ["A. keyboard", "B. mouse", "C. binder", "D. mousepad"], "answer": "D. mousepad", "target": "tissue box", "split": "train"}
{"id": "f248c2bcdc_18", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is near the mousepad, about 20 cm away. It is also near the keyboard, about 23 cm away. It is further near the mouse, about 24 cm away.", "target": "tissue box", "split": "train"}
{"id": "f248c2bcdc_19", "scene_id": "f248c2bcdc", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"near\", \"distance_cm\": 20.2}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 22.7}]}", "target": "tissue box", "split": "train"}
{"id": "f2e6c43543_0", "scene_id": "f2e6c43543", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "No", "target": "soap", "split": "train"}
{"id": "f2e6c43543_1", "scene_id": "f2e6c43543", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. plastic bottle\nC. table\nD. towel", "choices": ["A. chair", "B. plastic bottle", "C. table", "D. towel"], "answer": "A. chair", "target": "backpack", "split": "train"}
{"id": "f2e6c43543_2", "scene_id": "f2e6c43543", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the sofa. It is also next to the chair, about 6 cm away. It is further to the side of and above the towel, about 9 cm away.", "target": "backpack", "split": "train"}
{"id": "f2e6c43543_3", "scene_id": "f2e6c43543", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"sofa\", \"references\": [{\"object\": \"chair\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 9.4}]}", "target": "backpack", "split": "train"}
{"id": "f2ec1f6a04_0", "scene_id": "f2ec1f6a04", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "train"}
{"id": "f2ec1f6a04_1", "scene_id": "f2ec1f6a04", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. plates\nB. cutting boards\nC. mugs\nD. kitchen counter lamp", "choices": ["A. plates", "B. cutting boards", "C. mugs", "D. kitchen counter lamp"], "answer": "C. mugs", "target": "mug", "split": "train"}
{"id": "f2ec1f6a04_2", "scene_id": "f2ec1f6a04", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is below the mugs, about 11 cm away. It is also to the side of and above the kitchen counter lamp, about 40 cm away. It is further to the side of and above the plates, about 61 cm away.", "target": "mug", "split": "train"}
{"id": "f2ec1f6a04_3", "scene_id": "f2ec1f6a04", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"mugs\", \"relation\": \"below\", \"distance_cm\": 11.5}, {\"object\": \"kitchen counter lamp\", \"relation\": \"side_above\", \"distance_cm\": 39.7}]}", "target": "mug", "split": "train"}
{"id": "f368a1f730_0", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "f368a1f730_1", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. broom\nB. cooking appliances\nC. picture\nD. spray bottle", "choices": ["A. broom", "B. cooking appliances", "C. picture", "D. spray bottle"], "answer": "D. spray bottle", "target": "bottle", "split": "train"}
{"id": "f368a1f730_2", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is under the spray bottle, about 4 cm away. It is also to the side of and above the broom, about 14 cm away. It is further to the side of and below the picture, about 16 cm away.", "target": "bottle", "split": "train"}
{"id": "f368a1f730_3", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"spray bottle\", \"relation\": \"under\", \"distance_cm\": 4.4}, {\"object\": \"broom\", \"relation\": \"side_above\", \"distance_cm\": 13.6}]}", "target": "bottle", "split": "train"}
{"id": "f368a1f730_4", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "Yes", "target": "spray bottle", "split": "train"}
{"id": "f368a1f730_5", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the spray bottle in this scene?\nA. picture\nB. tray\nC. bottle\nD. broom", "choices": ["A. picture", "B. tray", "C. bottle", "D. broom"], "answer": "C. bottle", "target": "spray bottle", "split": "train"}
{"id": "f368a1f730_6", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the spray bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The spray bottle is to the side of and above the bottle, about 4 cm away. It is also to the side of and above the tray, about 8 cm away. It is further to the side of and below the picture, about 12 cm away.", "target": "spray bottle", "split": "train"}
{"id": "f368a1f730_7", "scene_id": "f368a1f730", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the spray bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"spray bottle\", \"support_surface\": \"storage rack\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_above\", \"distance_cm\": 4.4}, {\"object\": \"tray\", \"relation\": \"side_above\", \"distance_cm\": 8.2}]}", "target": "spray bottle", "split": "train"}
{"id": "f38b0108a1_0", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "f38b0108a1_1", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is above the carpet, about 30 cm away.", "target": "backpack", "split": "train"}
{"id": "f38b0108a1_2", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"carpet\", \"references\": [{\"object\": \"carpet\", \"relation\": \"above\", \"distance_cm\": 30.3}]}", "target": "backpack", "split": "train"}
{"id": "f38b0108a1_3", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "f38b0108a1_4", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. power socket\nB. kitchen bar\nC. tissue box\nD. kitchen utensils", "choices": ["A. power socket", "B. kitchen bar", "C. tissue box", "D. kitchen utensils"], "answer": "C. tissue box", "target": "cup", "split": "train"}
{"id": "f38b0108a1_5", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is to the side of and above the tissue box, about 25 cm away. It is also to the side of and above the kitchen bar, about 27 cm away. It is further to the side of and above the kitchen utensils, about 32 cm away.", "target": "cup", "split": "train"}
{"id": "f38b0108a1_6", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"food bag\", \"references\": [{\"object\": \"tissue box\", \"relation\": \"side_above\", \"distance_cm\": 25.2}, {\"object\": \"kitchen bar\", \"relation\": \"side_above\", \"distance_cm\": 26.6}]}", "target": "cup", "split": "train"}
{"id": "f38b0108a1_7", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "train"}
{"id": "f38b0108a1_8", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. monitor\nB. mouse\nC. rack\nD. keyboard", "choices": ["A. monitor", "B. mouse", "C. rack", "D. keyboard"], "answer": "A. monitor", "target": "headphones", "split": "train"}
{"id": "f38b0108a1_9", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the desk. It is also to the side of and below the monitor, about 6 cm away. It is further next to the keyboard, about 10 cm away.", "target": "headphones", "split": "train"}
{"id": "f38b0108a1_10", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.4}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.6}]}", "target": "headphones", "split": "train"}
{"id": "f38b0108a1_11", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "f38b0108a1_12", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. headphones\nB. office chair\nC. mouse\nD. thermos", "choices": ["A. headphones", "B. office chair", "C. mouse", "D. thermos"], "answer": "A. headphones", "target": "keyboard", "split": "train"}
{"id": "f38b0108a1_13", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the headphones, about 10 cm away. It is further to the side of and above the office chair, about 11 cm away.", "target": "keyboard", "split": "train"}
{"id": "f38b0108a1_14", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 9.6}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 11.2}]}", "target": "keyboard", "split": "train"}
{"id": "f38b0108a1_15", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "f38b0108a1_16", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. office chair\nB. headphones\nC. keyboard\nD. rack", "choices": ["A. office chair", "B. headphones", "C. keyboard", "D. rack"], "answer": "A. office chair", "target": "mouse", "split": "train"}
{"id": "f38b0108a1_17", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also to the side of and above the office chair, about 12 cm away. It is further next to the keyboard, about 14 cm away.", "target": "mouse", "split": "train"}
{"id": "f38b0108a1_18", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.4}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 13.7}]}", "target": "mouse", "split": "train"}
{"id": "f38b0108a1_19", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "f38b0108a1_20", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone in this scene?\nA. pillow\nB. power socket\nC. books\nD. mattress", "choices": ["A. pillow", "B. power socket", "C. books", "D. mattress"], "answer": "D. mattress", "target": "phone", "split": "train"}
{"id": "f38b0108a1_21", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone is on the bed. It is also to the side of and above the mattress, about 9 cm away. It is further to the side of and above the books, about 14 cm away.", "target": "phone", "split": "train"}
{"id": "f38b0108a1_22", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "22", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"mattress\", \"relation\": \"side_above\", \"distance_cm\": 8.8}, {\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 13.9}]}", "target": "phone", "split": "train"}
{"id": "f38b0108a1_23", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "23", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "f38b0108a1_24", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "24", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is next to the kitchen utensils, about 3 cm away. It is also under the kitchen cabinet, about 4 cm away. It is further to the side of and above the cutting board, about 13 cm away.", "target": "tissue box", "split": "train"}
{"id": "f38b0108a1_25", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen utensils\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"kitchen cabinet\", \"relation\": \"under\", \"distance_cm\": 3.8}]}", "target": "tissue box", "split": "train"}
{"id": "f38b0108a1_26", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "f38b0108a1_27", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the water bottle in this scene?\nA. mosquito swatter\nB. keyboard\nC. standing lamp\nD. desk mic", "choices": ["A. mosquito swatter", "B. keyboard", "C. standing lamp", "D. desk mic"], "answer": "A. mosquito swatter", "target": "water bottle", "split": "train"}
{"id": "f38b0108a1_28", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is on the desk. It is also to the side of and above the mosquito swatter, about 20 cm away. It is further to the side of and below the standing lamp, about 29 cm away.", "target": "water bottle", "split": "train"}
{"id": "f38b0108a1_29", "scene_id": "f38b0108a1", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mosquito swatter\", \"relation\": \"side_above\", \"distance_cm\": 20.2}, {\"object\": \"standing lamp\", \"relation\": \"side_below\", \"distance_cm\": 29.4}]}", "target": "water bottle", "split": "train"}
{"id": "f47ec86138_0", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "f47ec86138_1", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. split\nB. whiteboard marker\nC. trash can\nD. whiteboard", "choices": ["A. split", "B. whiteboard marker", "C. trash can", "D. whiteboard"], "answer": "A. split", "target": "bottle", "split": "train"}
{"id": "f47ec86138_2", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the split, about 6 cm away. It is also to the side of and below the whiteboard, about 9 cm away. It is further to the side of and above the trash can, about 47 cm away.", "target": "bottle", "split": "train"}
{"id": "f47ec86138_3", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"split\", \"references\": [{\"object\": \"split\", \"relation\": \"side_above\", \"distance_cm\": 6.3}, {\"object\": \"whiteboard\", \"relation\": \"side_below\", \"distance_cm\": 8.8}]}", "target": "bottle", "split": "train"}
{"id": "f47ec86138_4", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "train"}
{"id": "f47ec86138_5", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. pen\nB. monitor\nC. storage cabinet\nD. office chair", "choices": ["A. pen", "B. monitor", "C. storage cabinet", "D. office chair"], "answer": "B. monitor", "target": "mousepad", "split": "train"}
{"id": "f47ec86138_6", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is on the table. It is also near the pen, about 17 cm away. It is further to the side of and above the storage cabinet, about 29 cm away.", "target": "mousepad", "split": "train"}
{"id": "f47ec86138_7", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"pen\", \"relation\": \"near\", \"distance_cm\": 17.2}, {\"object\": \"storage cabinet\", \"relation\": \"side_above\", \"distance_cm\": 28.7}]}", "target": "mousepad", "split": "train"}
{"id": "f47ec86138_8", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "train"}
{"id": "f47ec86138_9", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notepad in this scene?\nA. monitor\nB. mouse\nC. computer tower\nD. office chair", "choices": ["A. monitor", "B. mouse", "C. computer tower", "D. office chair"], "answer": "D. office chair", "target": "notepad", "split": "train"}
{"id": "f47ec86138_10", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notepad is on the table. It is also to the side of and above the office chair, about 5 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "notepad", "split": "train"}
{"id": "f47ec86138_11", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"notepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 4.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.8}]}", "target": "notepad", "split": "train"}
{"id": "f47ec86138_12", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "train"}
{"id": "f47ec86138_13", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. mousepad\nB. monitor\nC. table\nD. storage cabinet", "choices": ["A. mousepad", "B. monitor", "C. table", "D. storage cabinet"], "answer": "C. table", "target": "pen", "split": "train"}
{"id": "f47ec86138_14", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is to the side of and above the table, about 9 cm away. It is also to the side of and above the storage cabinet, about 16 cm away. It is further to the side of and below the monitor, about 17 cm away.", "target": "pen", "split": "train"}
{"id": "f47ec86138_15", "scene_id": "f47ec86138", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"storage cabinet\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 8.6}, {\"object\": \"storage cabinet\", \"relation\": \"side_above\", \"distance_cm\": 15.6}]}", "target": "pen", "split": "train"}
{"id": "f4c5b0646f_0", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the extension cord appear?", "choices": null, "answer": "Yes", "target": "extension cord", "split": "train"}
{"id": "f4c5b0646f_1", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. coax outlet\nB. desk\nC. power sockets\nD. chair", "choices": ["A. coax outlet", "B. desk", "C. power sockets", "D. chair"], "answer": "C. power sockets", "target": "extension cord", "split": "train"}
{"id": "f4c5b0646f_2", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the power sockets, about 10 cm away. It is also to the side of and below the coax outlet, about 15 cm away. It is further to the side of and below the desk, about 21 cm away.", "target": "extension cord", "split": "train"}
{"id": "f4c5b0646f_3", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 10.4}, {\"object\": \"coax outlet\", \"relation\": \"side_below\", \"distance_cm\": 15.2}]}", "target": "extension cord", "split": "train"}
{"id": "f4c5b0646f_4", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "f4c5b0646f_5", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. pillow\nB. papers\nC. blanket\nD. book", "choices": ["A. pillow", "B. papers", "C. blanket", "D. book"], "answer": "B. papers", "target": "headphones", "split": "train"}
{"id": "f4c5b0646f_6", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is next to the papers, about 7 cm away. It is also to the side of and above the pillow, about 30 cm away. It is further to the side of and above the blanket, about 50 cm away.", "target": "headphones", "split": "train"}
{"id": "f4c5b0646f_7", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table lamp\", \"references\": [{\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 7.4}, {\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 29.8}]}", "target": "headphones", "split": "train"}
{"id": "f4c5b0646f_8", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "Yes", "target": "highlighter", "split": "train"}
{"id": "f4c5b0646f_9", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the highlighter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The highlighter is on the desk. It is also next to the keyboard, about 5 cm away. It is further to the side of and above the office chair, about 5 cm away.", "target": "highlighter", "split": "train"}
{"id": "f4c5b0646f_10", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the highlighter in structured JSON format.", "choices": null, "answer": "{\"target\": \"highlighter\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.7}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 5.2}]}", "target": "highlighter", "split": "train"}
{"id": "f4c5b0646f_11", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power adapter appear?", "choices": null, "answer": "No", "target": "power adapter", "split": "train"}
{"id": "f4c5b0646f_12", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also under the stand. It is further next to the pen.", "target": "keyboard", "split": "train"}
{"id": "f4c5b0646f_13", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"stand\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"pen\", \"relation\": \"next_to\", \"distance_cm\": 2.4}]}", "target": "keyboard", "split": "train"}
{"id": "f4c5b0646f_14", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "f4c5b0646f_15", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. trash can\nB. water bottle\nC. heater\nD. refillable water bottle", "choices": ["A. trash can", "B. water bottle", "C. heater", "D. refillable water bottle"], "answer": "D. refillable water bottle", "target": "notebook", "split": "train"}
{"id": "f4c5b0646f_16", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the table. It is also under the refillable water bottle, about 5 cm away. It is further to the side of and above the trash can, about 10 cm away.", "target": "notebook", "split": "train"}
{"id": "f4c5b0646f_17", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"refillable water bottle\", \"relation\": \"under\", \"distance_cm\": 4.9}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 9.7}]}", "target": "notebook", "split": "train"}
{"id": "f4c5b0646f_18", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote control appear?", "choices": null, "answer": "No", "target": "remote control", "split": "train"}
{"id": "f4c5b0646f_19", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen in this scene?\nA. mouse\nB. office chair\nC. clothes\nD. keyboard", "choices": ["A. mouse", "B. office chair", "C. clothes", "D. keyboard"], "answer": "D. keyboard", "target": "pen", "split": "train"}
{"id": "f4c5b0646f_20", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is on the desk. It is also next to the keyboard. It is further to the side of and above the office chair, about 7 cm away.", "target": "pen", "split": "train"}
{"id": "f4c5b0646f_21", "scene_id": "f4c5b0646f", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 7.0}]}", "target": "pen", "split": "train"}
{"id": "f5726eeeb2_0", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "train"}
{"id": "f5726eeeb2_1", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. chair\nB. bottle\nC. storage basket\nD. blanket", "choices": ["A. chair", "B. bottle", "C. storage basket", "D. blanket"], "answer": "D. blanket", "target": "backpack", "split": "train"}
{"id": "f5726eeeb2_2", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the blanket, about 7 cm away. It is also to the side of and below the bottle, about 40 cm away. It is further to the side of and above the storage basket, about 52 cm away.", "target": "backpack", "split": "train"}
{"id": "f5726eeeb2_3", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"blanket\", \"relation\": \"side_below\", \"distance_cm\": 6.8}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 40.0}]}", "target": "backpack", "split": "train"}
{"id": "f5726eeeb2_4", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the bottle appear?", "choices": null, "answer": "Yes", "target": "bottle", "split": "train"}
{"id": "f5726eeeb2_5", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. books\nB. storage basket\nC. blanket\nD. backpack", "choices": ["A. books", "B. storage basket", "C. blanket", "D. backpack"], "answer": "A. books", "target": "bottle", "split": "train"}
{"id": "f5726eeeb2_6", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the books, about 7 cm away. It is also to the side of and above the storage basket, about 26 cm away. It is further to the side of and above the backpack, about 40 cm away.", "target": "bottle", "split": "train"}
{"id": "f5726eeeb2_7", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 7.0}, {\"object\": \"storage basket\", \"relation\": \"side_above\", \"distance_cm\": 26.0}]}", "target": "bottle", "split": "train"}
{"id": "f5726eeeb2_8", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "f5726eeeb2_9", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. pillow\nB. bedside table\nC. bed\nD. backpack", "choices": ["A. pillow", "B. bedside table", "C. bed", "D. backpack"], "answer": "B. bedside table", "target": "phone charger", "split": "train"}
{"id": "f5726eeeb2_10", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and above the carpet, about 3 cm away. It is also to the side of and below the bedside table, about 16 cm away. It is further to the side of and below the pillow, about 52 cm away.", "target": "phone charger", "split": "train"}
{"id": "f5726eeeb2_11", "scene_id": "f5726eeeb2", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"carpet\", \"relation\": \"side_above\", \"distance_cm\": 3.1}, {\"object\": \"bedside table\", \"relation\": \"side_below\", \"distance_cm\": 16.5}]}", "target": "phone charger", "split": "train"}
{"id": "f576071590_0", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "f576071590_1", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. office chair\nB. monitor\nC. laptop\nD. mouse", "choices": ["A. office chair", "B. monitor", "C. laptop", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "train"}
{"id": "f576071590_2", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the desk. It is also next to the mouse, about 8 cm away. It is further to the side of and below the monitor, about 10 cm away.", "target": "keyboard", "split": "train"}
{"id": "f576071590_3", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 7.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.0}]}", "target": "keyboard", "split": "train"}
{"id": "f576071590_4", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "f576071590_5", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. office chair\nD. mousepad", "choices": ["A. keyboard", "B. monitor", "C. office chair", "D. mousepad"], "answer": "D. mousepad", "target": "mouse", "split": "train"}
{"id": "f576071590_6", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the mousepad, about 5 cm away. It is further next to the keyboard, about 8 cm away.", "target": "mouse", "split": "train"}
{"id": "f576071590_7", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 4.8}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.5}]}", "target": "mouse", "split": "train"}
{"id": "f576071590_8", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the stapler appear?", "choices": null, "answer": "No", "target": "stapler", "split": "train"}
{"id": "f576071590_9", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mousepad in this scene?\nA. mouse\nB. monitor\nC. office chair\nD. laptop", "choices": ["A. mouse", "B. monitor", "C. office chair", "D. laptop"], "answer": "B. monitor", "target": "mousepad", "split": "train"}
{"id": "f576071590_10", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is under the monitor, about 3 cm away. It is also next to the mouse, about 5 cm away. It is further to the side of and above the office chair, about 16 cm away.", "target": "mousepad", "split": "train"}
{"id": "f576071590_11", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.7}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.8}]}", "target": "mousepad", "split": "train"}
{"id": "f576071590_12", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "f576071590_13", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. foodstuffs\nB. paper towel roll\nC. pillow\nD. cardboard box", "choices": ["A. foodstuffs", "B. paper towel roll", "C. pillow", "D. cardboard box"], "answer": "A. foodstuffs", "target": "phone charger", "split": "train"}
{"id": "f576071590_14", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and below the foodstuffs, about 19 cm away. It is also to the side of and above the pillow, about 27 cm away. It is further near the paper towel roll, about 83 cm away.", "target": "phone charger", "split": "train"}
{"id": "f576071590_15", "scene_id": "f576071590", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"foodstuffs\", \"relation\": \"side_below\", \"distance_cm\": 18.8}, {\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 26.6}]}", "target": "phone charger", "split": "train"}
{"id": "f7d34e6706_0", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mousepad appear?", "choices": null, "answer": "No", "target": "mousepad", "split": "train"}
{"id": "f7d34e6706_1", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. hoodie\nB. cabinet\nC. office chair\nD. telephone", "choices": ["A. hoodie", "B. cabinet", "C. office chair", "D. telephone"], "answer": "B. cabinet", "target": "backpack", "split": "train"}
{"id": "f7d34e6706_2", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is under the cabinet, about 4 cm away. It is also to the side of and below the office chair, about 16 cm away. It is further to the side of and below the telephone, about 29 cm away.", "target": "backpack", "split": "train"}
{"id": "f7d34e6706_3", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"under\", \"distance_cm\": 3.6}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 16.4}]}", "target": "backpack", "split": "train"}
{"id": "f7d34e6706_4", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "f7d34e6706_5", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. document organizer\nB. book\nC. keyboard\nD. partition", "choices": ["A. document organizer", "B. book", "C. keyboard", "D. partition"], "answer": "A. document organizer", "target": "headphones", "split": "train"}
{"id": "f7d34e6706_6", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is on the table. It is also under the document organizer, about 3 cm away. It is further near the book, about 20 cm away.", "target": "headphones", "split": "train"}
{"id": "f7d34e6706_7", "scene_id": "f7d34e6706", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"document organizer\", \"relation\": \"under\", \"distance_cm\": 2.9}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 19.9}]}", "target": "headphones", "split": "train"}
{"id": "f7da3f0714_0", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the backpack appear?", "choices": null, "answer": "Yes", "target": "backpack", "split": "train"}
{"id": "f7da3f0714_1", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is to the side of and below the robe, about 44 cm away. It is also to the side of and below the chair, about 45 cm away. It is further to the side of and below the bottle, about 54 cm away.", "target": "backpack", "split": "train"}
{"id": "f7da3f0714_2", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"stool\", \"references\": [{\"object\": \"robe\", \"relation\": \"side_below\", \"distance_cm\": 44.1}, {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 44.9}]}", "target": "backpack", "split": "train"}
{"id": "f7da3f0714_3", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "train"}
{"id": "f7da3f0714_4", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. power adapter\nB. monitor\nC. office table\nD. power strip", "choices": ["A. power adapter", "B. monitor", "C. office table", "D. power strip"], "answer": "C. office table", "target": "cable", "split": "train"}
{"id": "f7da3f0714_5", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and below the office table, about 14 cm away. It is also to the side of and below the monitor, about 16 cm away. It is further to the side of and above the power strip, about 22 cm away.", "target": "cable", "split": "train"}
{"id": "f7da3f0714_6", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_below\", \"distance_cm\": 14.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 15.7}]}", "target": "cable", "split": "train"}
{"id": "f7da3f0714_7", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the spray bottle appear?", "choices": null, "answer": "No", "target": "spray bottle", "split": "train"}
{"id": "f7da3f0714_8", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse pad\nB. chair\nC. monitor\nD. mouse", "choices": ["A. mouse pad", "B. chair", "C. monitor", "D. mouse"], "answer": "A. mouse pad", "target": "keyboard", "split": "train"}
{"id": "f7da3f0714_9", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is next to the mouse pad, about 4 cm away. It is also next to the mouse, about 6 cm away. It is further to the side of and below the monitor, about 8 cm away.", "target": "keyboard", "split": "train"}
{"id": "f7da3f0714_10", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"mouse pad\", \"relation\": \"next_to\", \"distance_cm\": 4.1}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 6.5}]}", "target": "keyboard", "split": "train"}
{"id": "f7da3f0714_11", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the highlighter appear?", "choices": null, "answer": "No", "target": "highlighter", "split": "train"}
{"id": "f7da3f0714_12", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. chair\nC. office table\nD. pen holder", "choices": ["A. keyboard", "B. chair", "C. office table", "D. pen holder"], "answer": "C. office table", "target": "mouse", "split": "train"}
{"id": "f7da3f0714_13", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and above the office table, about 3 cm away. It is also next to the keyboard, about 6 cm away. It is further to the side of and above the chair, about 9 cm away.", "target": "mouse", "split": "train"}
{"id": "f7da3f0714_14", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 2.7}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 6.5}]}", "target": "mouse", "split": "train"}
{"id": "f7da3f0714_15", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen appear?", "choices": null, "answer": "No", "target": "pen", "split": "train"}
{"id": "f7da3f0714_16", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "16", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is under the pen holder, about 3 cm away. It is also next to the keyboard, about 4 cm away. It is further next to the robe, about 8 cm away.", "target": "mouse pad", "split": "train"}
{"id": "f7da3f0714_17", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "17", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"pen holder\", \"relation\": \"under\", \"distance_cm\": 3.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.1}]}", "target": "mouse pad", "split": "train"}
{"id": "f7da3f0714_18", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "18", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "train"}
{"id": "f7da3f0714_19", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "19", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power adapter in this scene?\nA. cable\nB. chair\nC. office table\nD. power socket", "choices": ["A. cable", "B. chair", "C. office table", "D. power socket"], "answer": "A. cable", "target": "power adapter", "split": "train"}
{"id": "f7da3f0714_20", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "20", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power adapter in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power adapter is to the side of and below the cable, about 24 cm away. It is also to the side of and below the office table, about 35 cm away. It is further to the side of and below the chair, about 42 cm away.", "target": "power adapter", "split": "train"}
{"id": "f7da3f0714_21", "scene_id": "f7da3f0714", "source_dataset": "scannetpp", "local_sample_id": "21", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power adapter in structured JSON format.", "choices": null, "answer": "{\"target\": \"power adapter\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"cable\", \"relation\": \"side_below\", \"distance_cm\": 24.3}, {\"object\": \"office table\", \"relation\": \"side_below\", \"distance_cm\": 35.3}]}", "target": "power adapter", "split": "train"}
{"id": "f8e13ab4ae_0", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "train"}
{"id": "f8e13ab4ae_1", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the backpack in this scene?\nA. frying pan\nB. box\nC. cleaning agent bottle\nD. food container", "choices": ["A. frying pan", "B. box", "C. cleaning agent bottle", "D. food container"], "answer": "A. frying pan", "target": "backpack", "split": "train"}
{"id": "f8e13ab4ae_2", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the backpack in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The backpack is on the bookshelf. It is also to the side of and above the frying pan, about 21 cm away. It is further to the side of and above the food container, about 25 cm away.", "target": "backpack", "split": "train"}
{"id": "f8e13ab4ae_3", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the backpack in structured JSON format.", "choices": null, "answer": "{\"target\": \"backpack\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"frying pan\", \"relation\": \"side_above\", \"distance_cm\": 20.5}, {\"object\": \"food container\", \"relation\": \"side_above\", \"distance_cm\": 25.0}]}", "target": "backpack", "split": "train"}
{"id": "f8e13ab4ae_4", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the dongle appear?", "choices": null, "answer": "No", "target": "dongle", "split": "train"}
{"id": "f8e13ab4ae_5", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the laptop charger in this scene?\nA. box\nB. phone charger\nC. rj45 outlet\nD. coax outlet", "choices": ["A. box", "B. phone charger", "C. rj45 outlet", "D. coax outlet"], "answer": "B. phone charger", "target": "laptop charger", "split": "train"}
{"id": "f8e13ab4ae_6", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is above the rj45 outlet, about 12 cm away. It is also above the coax outlet, about 24 cm away. It is further to the side of and below the box, about 59 cm away.", "target": "laptop charger", "split": "train"}
{"id": "f8e13ab4ae_7", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"box\", \"references\": [{\"object\": \"rj45 outlet\", \"relation\": \"above\", \"distance_cm\": 12.1}, {\"object\": \"coax outlet\", \"relation\": \"above\", \"distance_cm\": 23.6}]}", "target": "laptop charger", "split": "train"}
{"id": "f8e13ab4ae_8", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "train"}
{"id": "f8e13ab4ae_9", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the notebook in this scene?\nA. coax outlet\nB. laptop charger\nC. rj45 outlet\nD. box", "choices": ["A. coax outlet", "B. laptop charger", "C. rj45 outlet", "D. box"], "answer": "D. box", "target": "notebook", "split": "train"}
{"id": "f8e13ab4ae_10", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is below the box, about 60 cm away. It is also near the coax outlet, about 62 cm away. It is further to the side of and below the rj45 outlet, about 64 cm away.", "target": "notebook", "split": "train"}
{"id": "f8e13ab4ae_11", "scene_id": "f8e13ab4ae", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"bookshelf\", \"references\": [{\"object\": \"box\", \"relation\": \"below\", \"distance_cm\": 60.4}, {\"object\": \"coax outlet\", \"relation\": \"near\", \"distance_cm\": 62.4}]}", "target": "notebook", "split": "train"}
{"id": "f9397af4cb_0", "scene_id": "f9397af4cb", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "train"}
{"id": "f9397af4cb_1", "scene_id": "f9397af4cb", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. pinboard\nB. chair\nC. laptop\nD. power socket", "choices": ["A. pinboard", "B. chair", "C. laptop", "D. power socket"], "answer": "B. chair", "target": "power strip", "split": "train"}
{"id": "f9397af4cb_2", "scene_id": "f9397af4cb", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the chair, about 18 cm away. It is also to the side of and below the power socket, about 30 cm away. It is further to the side of and below the pinboard, about 53 cm away.", "target": "power strip", "split": "train"}
{"id": "f9397af4cb_3", "scene_id": "f9397af4cb", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 18.5}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 30.5}]}", "target": "power strip", "split": "train"}
{"id": "f97de2c3e9_0", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the glasses case appear?", "choices": null, "answer": "No", "target": "glasses case", "split": "train"}
{"id": "f97de2c3e9_1", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. kitchen cabinet\nB. suitcase\nC. microwave\nD. table", "choices": ["A. kitchen cabinet", "B. suitcase", "C. microwave", "D. table"], "answer": "D. table", "target": "bottle", "split": "train"}
{"id": "f97de2c3e9_2", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and below the table, about 7 cm away. It is also to the side of and below the suitcase, about 11 cm away. It is further to the side of and below the microwave, about 20 cm away.", "target": "bottle", "split": "train"}
{"id": "f97de2c3e9_3", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 6.8}, {\"object\": \"suitcase\", \"relation\": \"side_below\", \"distance_cm\": 11.3}]}", "target": "bottle", "split": "train"}
{"id": "f97de2c3e9_4", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "f97de2c3e9_5", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. laptop\nB. curtain\nC. chair\nD. laptop cover", "choices": ["A. laptop", "B. curtain", "C. chair", "D. laptop cover"], "answer": "A. laptop", "target": "mouse", "split": "train"}
{"id": "f97de2c3e9_6", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the desk. It is also next to the laptop, about 7 cm away. It is further to the side of and below the laptop cover, about 10 cm away.", "target": "mouse", "split": "train"}
{"id": "f97de2c3e9_7", "scene_id": "f97de2c3e9", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"laptop cover\", \"relation\": \"side_below\", \"distance_cm\": 10.3}]}", "target": "mouse", "split": "train"}
{"id": "f9f95681fd_0", "scene_id": "f9f95681fd", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the remote appear?", "choices": null, "answer": "No", "target": "remote", "split": "train"}
{"id": "f9f95681fd_1", "scene_id": "f9f95681fd", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. kettle\nB. knife set\nC. cheese\nD. chopping board", "choices": ["A. kettle", "B. knife set", "C. cheese", "D. chopping board"], "answer": "C. cheese", "target": "bottle", "split": "train"}
{"id": "f9f95681fd_2", "scene_id": "f9f95681fd", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is next to the cheese, about 7 cm away. It is also next to the knife set, about 12 cm away. It is further to the side of and above the chopping board, about 28 cm away.", "target": "bottle", "split": "train"}
{"id": "f9f95681fd_3", "scene_id": "f9f95681fd", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"cheese\", \"relation\": \"next_to\", \"distance_cm\": 7.2}, {\"object\": \"knife set\", \"relation\": \"next_to\", \"distance_cm\": 11.5}]}", "target": "bottle", "split": "train"}
{"id": "faec2f0468_0", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "train"}
{"id": "faec2f0468_1", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. light switch\nB. wall lamp\nC. box\nD. book", "choices": ["A. light switch", "B. wall lamp", "C. box", "D. book"], "answer": "D. book", "target": "bottle", "split": "train"}
{"id": "faec2f0468_2", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the book, about 5 cm away. It is also to the side of and below the wall lamp, about 69 cm away. It is further to the side of and above the box, about 72 cm away.", "target": "bottle", "split": "train"}
{"id": "faec2f0468_3", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"wall lamp\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 5.5}, {\"object\": \"wall lamp\", \"relation\": \"side_below\", \"distance_cm\": 68.5}]}", "target": "bottle", "split": "train"}
{"id": "faec2f0468_4", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "train"}
{"id": "faec2f0468_5", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. mouse\nB. photo frame\nC. monitor\nD. office chair", "choices": ["A. mouse", "B. photo frame", "C. monitor", "D. office chair"], "answer": "C. monitor", "target": "keyboard", "split": "train"}
{"id": "faec2f0468_6", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is on the table. It is also to the side of and below the monitor, about 6 cm away. It is further next to the mouse, about 13 cm away.", "target": "keyboard", "split": "train"}
{"id": "faec2f0468_7", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.7}]}", "target": "keyboard", "split": "train"}
{"id": "faec2f0468_8", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the wireless charger appear?", "choices": null, "answer": "No", "target": "wireless charger", "split": "train"}
{"id": "faec2f0468_9", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. keyboard\nB. monitor\nC. photo frame\nD. office chair", "choices": ["A. keyboard", "B. monitor", "C. photo frame", "D. office chair"], "answer": "A. keyboard", "target": "mouse", "split": "train"}
{"id": "faec2f0468_10", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the keyboard, about 13 cm away. It is further to the side of and above the office chair, about 17 cm away.", "target": "mouse", "split": "train"}
{"id": "faec2f0468_11", "scene_id": "faec2f0468", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 12.7}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 16.8}]}", "target": "mouse", "split": "train"}
{"id": "fb152519ad_0", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the post it note appear?", "choices": null, "answer": "No", "target": "post it note", "split": "train"}
{"id": "fb152519ad_1", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. shelf\nB. paper\nC. coffee maker\nD. kettle", "choices": ["A. shelf", "B. paper", "C. coffee maker", "D. kettle"], "answer": "C. coffee maker", "target": "bottle", "split": "train"}
{"id": "fb152519ad_2", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also to the side of and above the coffee maker, about 4 cm away. It is further next to the kettle, about 5 cm away.", "target": "bottle", "split": "train"}
{"id": "fb152519ad_3", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"coffee maker\", \"relation\": \"side_above\", \"distance_cm\": 4.2}, {\"object\": \"kettle\", \"relation\": \"next_to\", \"distance_cm\": 5.3}]}", "target": "bottle", "split": "train"}
{"id": "fb152519ad_4", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the soap appear?", "choices": null, "answer": "Yes", "target": "soap", "split": "train"}
{"id": "fb152519ad_5", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the soap in this scene?\nA. shower wall\nB. toiletry\nC. towel\nD. shower rack", "choices": ["A. shower wall", "B. toiletry", "C. towel", "D. shower rack"], "answer": "A. shower wall", "target": "soap", "split": "train"}
{"id": "fb152519ad_6", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the soap in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The soap is to the side of and below the shower wall, about 6 cm away. It is also to the side of and below the toiletry, about 23 cm away. It is further to the side of and below the towel, about 28 cm away.", "target": "soap", "split": "train"}
{"id": "fb152519ad_7", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the soap in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap\", \"support_surface\": \"shower rack\", \"references\": [{\"object\": \"shower wall\", \"relation\": \"side_below\", \"distance_cm\": 5.5}, {\"object\": \"toiletry\", \"relation\": \"side_below\", \"distance_cm\": 23.2}]}", "target": "soap", "split": "train"}
{"id": "fb152519ad_8", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "fb152519ad_9", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the tissue box in this scene?\nA. paper towel\nB. packet\nC. speaker\nD. box", "choices": ["A. paper towel", "B. packet", "C. speaker", "D. box"], "answer": "D. box", "target": "tissue box", "split": "train"}
{"id": "fb152519ad_10", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is on the shelf. It is also to the side of and above the box, about 4 cm away. It is further next to the packet, about 8 cm away.", "target": "tissue box", "split": "train"}
{"id": "fb152519ad_11", "scene_id": "fb152519ad", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 4.3}, {\"object\": \"packet\", \"relation\": \"next_to\", \"distance_cm\": 8.1}]}", "target": "tissue box", "split": "train"}
{"id": "fb5a96b1a2_0", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "train"}
{"id": "fb5a96b1a2_1", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. chair\nB. air conditioner\nC. bottle spray\nD. box", "choices": ["A. chair", "B. air conditioner", "C. bottle spray", "D. box"], "answer": "C. bottle spray", "target": "bottle", "split": "train"}
{"id": "fb5a96b1a2_2", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the table. It is also next to the bottle spray, about 3 cm away. It is further to the side of and above the chair, about 42 cm away.", "target": "bottle", "split": "train"}
{"id": "fb5a96b1a2_3", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bottle spray\", \"relation\": \"next_to\", \"distance_cm\": 2.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 42.1}]}", "target": "bottle", "split": "train"}
{"id": "fb5a96b1a2_4", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "fb5a96b1a2_5", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. picture\nB. chair\nC. box\nD. book", "choices": ["A. picture", "B. chair", "C. box", "D. book"], "answer": "A. picture", "target": "mouse", "split": "train"}
{"id": "fb5a96b1a2_6", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is on the table. It is also next to the picture, about 5 cm away. It is further to the side of and above the chair, about 27 cm away.", "target": "mouse", "split": "train"}
{"id": "fb5a96b1a2_7", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"picture\", \"relation\": \"next_to\", \"distance_cm\": 5.4}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 27.4}]}", "target": "mouse", "split": "train"}
{"id": "fb5a96b1a2_8", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "No", "target": "cup", "split": "train"}
{"id": "fb5a96b1a2_9", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the power strip in this scene?\nA. power socket\nB. air conditioner\nC. table\nD. sofa", "choices": ["A. power socket", "B. air conditioner", "C. table", "D. sofa"], "answer": "C. table", "target": "power strip", "split": "train"}
{"id": "fb5a96b1a2_10", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the power strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is to the side of and below the table, about 15 cm away. It is also to the side of and below the air conditioner, about 35 cm away. It is further to the side of and below the sofa, about 37 cm away.", "target": "power strip", "split": "train"}
{"id": "fb5a96b1a2_11", "scene_id": "fb5a96b1a2", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 15.2}, {\"object\": \"air conditioner\", \"relation\": \"side_below\", \"distance_cm\": 35.3}]}", "target": "power strip", "split": "train"}
{"id": "fb893ffaf3_0", "scene_id": "fb893ffaf3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "train"}
{"id": "fb893ffaf3_1", "scene_id": "fb893ffaf3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"shelf\", \"references\": []}", "target": "remote", "split": "train"}
{"id": "fb9b4c2f15_0", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power bank appear?", "choices": null, "answer": "No", "target": "power bank", "split": "train"}
{"id": "fb9b4c2f15_1", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. backpack\nB. shelf rail\nC. shelf\nD. storage box", "choices": ["A. backpack", "B. shelf rail", "C. shelf", "D. storage box"], "answer": "C. shelf", "target": "bottle", "split": "train"}
{"id": "fb9b4c2f15_2", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is on the shelf. It is also to the side of and above the shelf rail, about 12 cm away. It is further to the side of and above the storage box, about 14 cm away.", "target": "bottle", "split": "train"}
{"id": "fb9b4c2f15_3", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_above\", \"distance_cm\": 11.6}, {\"object\": \"storage box\", \"relation\": \"side_above\", \"distance_cm\": 13.5}]}", "target": "bottle", "split": "train"}
{"id": "fb9b4c2f15_4", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone stand appear?", "choices": null, "answer": "No", "target": "phone stand", "split": "train"}
{"id": "fb9b4c2f15_5", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the extension cord in this scene?\nA. power outlets\nB. thermos\nC. phone charger\nD. shelf rail", "choices": ["A. power outlets", "B. thermos", "C. phone charger", "D. shelf rail"], "answer": "D. shelf rail", "target": "extension cord", "split": "train"}
{"id": "fb9b4c2f15_6", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the extension cord in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The extension cord is to the side of and below the shelf rail, about 7 cm away. It is also to the side of and below the thermos, about 26 cm away. It is further to the side of and above the power outlets, about 32 cm away.", "target": "extension cord", "split": "train"}
{"id": "fb9b4c2f15_7", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the extension cord in structured JSON format.", "choices": null, "answer": "{\"target\": \"extension cord\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_below\", \"distance_cm\": 6.5}, {\"object\": \"thermos\", \"relation\": \"side_below\", \"distance_cm\": 26.0}]}", "target": "extension cord", "split": "train"}
{"id": "fb9b4c2f15_8", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the phone charger appear?", "choices": null, "answer": "Yes", "target": "phone charger", "split": "train"}
{"id": "fb9b4c2f15_9", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the phone charger in this scene?\nA. laptop bag\nB. router\nC. jacket\nD. backpack", "choices": ["A. laptop bag", "B. router", "C. jacket", "D. backpack"], "answer": "C. jacket", "target": "phone charger", "split": "train"}
{"id": "fb9b4c2f15_10", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the phone charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone charger is to the side of and above the jacket, about 31 cm away. It is also to the side of and below the laptop bag, about 45 cm away. It is further to the side of and below the router, about 79 cm away.", "target": "phone charger", "split": "train"}
{"id": "fb9b4c2f15_11", "scene_id": "fb9b4c2f15", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"table\", \"references\": [{\"object\": \"jacket\", \"relation\": \"side_above\", \"distance_cm\": 30.6}, {\"object\": \"laptop bag\", \"relation\": \"side_below\", \"distance_cm\": 44.5}]}", "target": "phone charger", "split": "train"}
{"id": "fd361ab85f_0", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the earbuds appear?", "choices": null, "answer": "No", "target": "earbuds", "split": "train"}
{"id": "fd361ab85f_1", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cable in this scene?\nA. box\nB. bottle\nC. storage cabinet\nD. tv", "choices": ["A. box", "B. bottle", "C. storage cabinet", "D. tv"], "answer": "D. tv", "target": "cable", "split": "train"}
{"id": "fd361ab85f_2", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is to the side of and below the tv, about 8 cm away. It is also to the side of and below the storage cabinet, about 38 cm away.", "target": "cable", "split": "train"}
{"id": "fd361ab85f_3", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"mobile tv stand\", \"references\": [{\"object\": \"tv\", \"relation\": \"side_below\", \"distance_cm\": 7.8}, {\"object\": \"storage cabinet\", \"relation\": \"side_below\", \"distance_cm\": 37.5}]}", "target": "cable", "split": "train"}
{"id": "fd361ab85f_4", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "train"}
{"id": "fd361ab85f_5", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mug in this scene?\nA. monitor\nB. office chair\nC. window sill\nD. mouse", "choices": ["A. monitor", "B. office chair", "C. window sill", "D. mouse"], "answer": "D. mouse", "target": "mug", "split": "train"}
{"id": "fd361ab85f_6", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mug in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mug is on the table. It is also next to the mouse, about 9 cm away. It is further to the side of and below the monitor, about 27 cm away.", "target": "mug", "split": "train"}
{"id": "fd361ab85f_7", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 8.8}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 27.4}]}", "target": "mug", "split": "train"}
{"id": "fd361ab85f_8", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the power plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "train"}
{"id": "fd361ab85f_9", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the remote in this scene?\nA. paper\nB. keyboard\nC. screen\nD. office chair", "choices": ["A. paper", "B. keyboard", "C. screen", "D. office chair"], "answer": "D. office chair", "target": "remote", "split": "train"}
{"id": "fd361ab85f_10", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote is on the table. It is also to the side of and above the office chair, about 6 cm away. It is further near the keyboard, about 22 cm away.", "target": "remote", "split": "train"}
{"id": "fd361ab85f_11", "scene_id": "fd361ab85f", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote\", \"support_surface\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 21.8}]}", "target": "remote", "split": "train"}
{"id": "fdf7f5cd5d_0", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the tissue appear?", "choices": null, "answer": "No", "target": "tissue", "split": "train"}
{"id": "fdf7f5cd5d_1", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the bottle in this scene?\nA. plant\nB. stapler\nC. book\nD. plant pot", "choices": ["A. plant", "B. stapler", "C. book", "D. plant pot"], "answer": "C. book", "target": "bottle", "split": "train"}
{"id": "fdf7f5cd5d_2", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bottle is to the side of and above the book, about 8 cm away. It is also to the side of and below the plant, about 38 cm away. It is further near the stapler, about 39 cm away.", "target": "bottle", "split": "train"}
{"id": "fdf7f5cd5d_3", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"book\", \"references\": [{\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 8.5}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 38.1}]}", "target": "bottle", "split": "train"}
{"id": "fdf7f5cd5d_4", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the cup appear?", "choices": null, "answer": "Yes", "target": "cup", "split": "train"}
{"id": "fdf7f5cd5d_5", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the cup in this scene?\nA. box\nB. book\nC. hole puncher\nD. plant pot", "choices": ["A. box", "B. book", "C. hole puncher", "D. plant pot"], "answer": "C. hole puncher", "target": "cup", "split": "train"}
{"id": "fdf7f5cd5d_6", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the cup in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cup is next to the hole puncher, about 5 cm away. It is also to the side of and below the plant pot, about 7 cm away. It is further next to the box, about 9 cm away.", "target": "cup", "split": "train"}
{"id": "fdf7f5cd5d_7", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"snack bag\", \"references\": [{\"object\": \"hole puncher\", \"relation\": \"next_to\", \"distance_cm\": 4.7}, {\"object\": \"plant pot\", \"relation\": \"side_below\", \"distance_cm\": 6.8}]}", "target": "cup", "split": "train"}
{"id": "fdf7f5cd5d_8", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "8", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the headphones appear?", "choices": null, "answer": "Yes", "target": "headphones", "split": "train"}
{"id": "fdf7f5cd5d_9", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "9", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the headphones in this scene?\nA. books\nB. laptop\nC. monitor\nD. cabinet", "choices": ["A. books", "B. laptop", "C. monitor", "D. cabinet"], "answer": "C. monitor", "target": "headphones", "split": "train"}
{"id": "fdf7f5cd5d_10", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "10", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the headphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headphones is to the side of and below the monitor, about 10 cm away. It is also next to the laptop, about 11 cm away. It is further to the side of and above the cabinet, about 29 cm away.", "target": "headphones", "split": "train"}
{"id": "fdf7f5cd5d_11", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "11", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.7}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 11.2}]}", "target": "headphones", "split": "train"}
{"id": "fdf7f5cd5d_12", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "12", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the marker appear?", "choices": null, "answer": "No", "target": "marker", "split": "train"}
{"id": "fdf7f5cd5d_13", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "13", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the pen holder in this scene?\nA. book\nB. plant pot\nC. stapler\nD. plant", "choices": ["A. book", "B. plant pot", "C. stapler", "D. plant"], "answer": "B. plant pot", "target": "pen holder", "split": "train"}
{"id": "fdf7f5cd5d_14", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "14", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the pen holder in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen holder is to the side of and below the plant pot, about 16 cm away. It is also to the side of and below the plant, about 17 cm away. It is further to the side of and above the book, about 19 cm away.", "target": "pen holder", "split": "train"}
{"id": "fdf7f5cd5d_15", "scene_id": "fdf7f5cd5d", "source_dataset": "scannetpp", "local_sample_id": "15", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"box\", \"references\": [{\"object\": \"plant pot\", \"relation\": \"side_below\", \"distance_cm\": 16.1}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 17.4}]}", "target": "pen holder", "split": "train"}
{"id": "fe5fe0a8a4_0", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "train"}
{"id": "fe5fe0a8a4_1", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the keyboard in this scene?\nA. cable\nB. file cabinet\nC. extension cord\nD. monitor", "choices": ["A. cable", "B. file cabinet", "C. extension cord", "D. monitor"], "answer": "B. file cabinet", "target": "keyboard", "split": "train"}
{"id": "fe5fe0a8a4_2", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the keyboard in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The keyboard is to the side of and above the file cabinet, about 5 cm away.", "target": "keyboard", "split": "train"}
{"id": "fe5fe0a8a4_3", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"file cabinet\", \"references\": [{\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 5.0}]}", "target": "keyboard", "split": "train"}
{"id": "fe5fe0a8a4_4", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "4", "task": "TPV", "question_type": "target presence verification", "instruction": "Answer only with exactly one word: Yes or No. Do not add any explanation.", "question": "In the entire scene, did the mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "train"}
{"id": "fe5fe0a8a4_5", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "5", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output any other text.", "question": "Excluding the surface it is resting on, which of the following objects is nearest to the final location of the mouse in this scene?\nA. monitor\nB. pc\nC. file cabinet\nD. cable", "choices": ["A. monitor", "B. pc", "C. file cabinet", "D. cable"], "answer": "A. monitor", "target": "mouse", "split": "train"}
{"id": "fe5fe0a8a4_6", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "6", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). Do not use meters or vague approximations. Use everyday English and varied sentence structures.", "question": "Please describe in detail the final location of the mouse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse is to the side of and below the monitor, about 6 cm away. It is also to the side of and above the file cabinet, about 15 cm away. It is further to the side of and below the pc, about 44 cm away.", "target": "mouse", "split": "train"}
{"id": "fe5fe0a8a4_7", "scene_id": "fe5fe0a8a4", "source_dataset": "scannetpp", "local_sample_id": "7", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm must be float (add .0 if integer)\n3. JSON must be directly parsable with json.loads()", "question": "Please output the final location of the mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.5}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 15.1}]}", "target": "mouse", "split": "train"}